The Habitat of the Eurypterida/Chapter IV
CHAPTER IV
The Lithogenesis of the Euryptertd-bearing Beds
The formations which in America contain eurypterids in abundance are:
- The Belt Terrane.
- The Normanskill and Schenectady beds.
- The Shawangunk conglomerate.
- The Pittsford shale.
- The Bertie waterlime.
- The Kokomo waterlimes.
Those most prolific in Europe are:
- The Tarannon and Wenlock beds of southern Scotland.
- The waterlime beds of Oesel.
- The Siluric of the Austro-Russian border lands.
- The Ludlow of England and Ludlow and Lanarkian of Scotland.
- The Old Red Sandstone.
It is evident that the formations carrying only fragments or single individuals need not be considered if we can prove a uniform habitat from the formations carrying these merostomes in abundance. Nevertheless a brief summary of these is also given at the end of the chapter.
1. THE BELT TERRANE
The Belt Terrane fauna is a large one made up of fragments which Clarke and Ruedemann have failed to identify as of merostome origin, though Walcott insists that they belong to this group. Clarke and Ruedemann hold, however, that the specimens from the Altyn limestone of Alberta are undoubtedly merostomes, but they question the correlation of the Altyn and Belt Terrane, and the consequent reference of the remains from the two formations to Beltina danai. (This has been discussed on pp. 11-13.) The Belt Terrane material, nevertheless, has a very strong resemblance to the eurypterid fragments from other horizons, though the specimens all lack the surface markings characteristic of the eurypterids. Some of the most typical material is figured on plate 25 in the Bulletin of the Geological Society of America, Vol. X, 1898, and again in the Smithsonian Miscellaneous Collections, Vol. LXIV, No. 2, plate 22.
Of the conditions of sedimentation prevailing during this period Walcott says:
"Briefly summarized, the Algonkian period in North America with its great epicontinental formations was a time of continental elevation and largely terrigerious sedimentation in non-marine bodies of water, and of deposition by aerial and stream processes in favorable areas. . . . .
"The North American continent was larger at the close of Algonkian time than at any subsequent period other than possibly at the end of the Cretaceous, when the land was equally extensive. Indeed, it is highly probable that its area was greater then than even now, for no marine deposits of Algonkian age containing pre-Cambrian life, as they were laid down in Lipalian[1] time immediately preceding the Cambrian period have been discovered on the North American continent or elsewhere, so far as known" (290, 81, 82).
Walcott does not wholly subscribe to the fresh water habitat of these eurypterids early for he speaks of Beltina danai as "possibly of marine derivation" and uses the presence of this fossil in the Belt Terrane as an indication that a connection of the Cordilleran geosyncline with the sea was temporarily effected allowing "at least a crustacean, and a few annelids" to become adapted to the Montana-Alberta sea. It is clear that Walcott allows this entrance of the sea into the Beltian lake only in order to account for the presence of the eurypterids and annelids and to conform to the prevailing opinion that the early eurypterids were marine organisms. This concession seems to be unnecessary. The annelids surely would be more naturally accounted for as terrestrial forms and besides, it would scarcely be possible for them to leave their trails in deposits formed under water. Such trails would have to be made on surfaces exposed to the air long enough to harden and to be covered by wind-blown sand or dust or by a fresh deposit of water-laid material, but in the latter case a sufficient length of time would have to elapse to allow of the thorough hardening of the trail. In this case as in many another the question must be raised: Why if the eurypterids were marine were they the only organisms which were carried in from the open sea? It is well known that the littoral waters of the Pre-Cambric must have teamed with all the forms of life which are so abundantly represented in the advancing Cambric waters. It seems absurd to suppose that thousands of fragments of a eurypterid should have been washed in from the sea, but no other marine form.
The great thicknesses of Algonkian limestone found in the Belt terrane and corresponding formations have been adequately accounted for by Walcott as algal deposits in a series of lakes formed within the Cordilleran geosyncline. "The lakes of Algonkian [Pre-Cambric[2]] time were not much if any larger in area than the 'Great Lakes' of the St. Lawrence drainage basin and they were much shallower and more laden with mud and mineral matter in solution.
"The area of the Belt terrane in Montana is about 6000 square miles. This seems large when studying it in the field, but it is only one-fifth of the size of our great fresh-water Lake Superior" (290, 89).
Walcott has described nine species of calcareous algae from the Newland limestone below the Greyson shales and one which is abundant in the Spokane shales just above the Greyson. It is much more logical to suppose that the Greyson shales represent river rather than marine deposits, for they are coarse and arenaceous with interbedded shales, in which algal reefs could not grow. This would account for the absence of the reefs in the Greyson and for the absence of the eurypterids in the Newland limestone. I make this suggestion merely as a more plausible explanation of the conditions than the one which is usually offered.
If it can be assumed as proved that the remains in the Belt Terrane are of eurypterid affinity, they would offer just the proof desired to show that the eurypterids from the earliest times lived in terrestrial waters, for the Belt terrane has been shown by Barrell from purely lithological evidence to be non-marine. From the presence of mud-cracks and other structural characters of the formation he concludes that the terrane gives evidence of "two sedimentary cycles, each of which contains a strongly marked formation of mudcracked red shales, the shales alternating with sandy strata, and both judged to have been deposited on the flood plains of rivers, whose deltas had gained over the subsidence, finally filling up and displacing the shallow epicontinental sea" (14, 319, 320).
2. THE NORMANSKILL AND SCHENECTADY BEDS
The Normanskill sandstones and shales of Catskill and the Schenectady bluestones of Schenectady, New York, are so similar lithologically and faunally that they may be considered together. Comparatively little is known concerning the details of distribution of these formations and their physical changes from place to place, yet the descriptions available and the studies I have been able to make in the field, make clear what must be the origin of the sediments. In reference to the Normanskill beds Clarke and Ruedemann make the following statement:
"The lithologic and faunal conditions at the Broom street quarry exposure were found to be a singularly complete duplication of those of the eurypterid-bearing exposures in the bluestone quarries at Schenectady. The Broom street quarry is also a bluestone quarry, the rock being mostly used in the crusher. The courses of 'bluestone' (here an impure argillaceous sandstone) are very compact, from 3 to 30 feet thick between the intercalations of black shales. There is distinct evidence of shallow water conditions, one bed being conglomeratic and largely composed of pebbles, many of which appear to be mud pebbles; another beautifully exhibiting very regular, widely separated wave marks with winnows of comminuted seaweeds and eurypterids in the troughs.
"Quite as in the bluestone quarries of the Schenectady beds, the surfaces of some of the sandstones are densely covered with rather poorly preserved seaweeds and eurypterids. It was therefore natural to expect that here too the black intercalated shales would contain better material of these fossils and possibly also graptolites that would indicate the age of the beds. They have indeed afforded a layer with an association of finely preserved seaweeds, the eurypterids herewith described, and the following graptolites: Dicellograptus gurleyi Lapworth, Climacograptus bicornis Hall, Climacograptus bicornis var. peltifer Lapworth, Cryptograptus tricornis (Carruthers), the first three forms in great abundance. This graptolite association is one of undoubted Normanskill age. The seaweeds occur in large perfect fronds and are of the same type as those in the Schenectady shale. The eurypterids also are strikingly similar to those from the Schenectady beds" (39, 411, 412).
The eurypterid remains are very fragmentary, in fact, they are so incomplete that generic determinations are only provisional, there being but a few carapaces and fragmentary abdomina with a small number of legs and telsons rarely attached. Five genera are thought to be represented: Eurypterus, Eusarcus, Dolichopterus, Stylonurus by one species each, and Pterygotus by two species, though one of these may be a Eusarcus.
The physical characteristics of the Schenectady beds are closely similar to those of the Normanskill beds. Both consist of heavy bedded sandstones, dark in color, but highly siliceous, alternating with black shales. The sandstones are compact enough to be quarried for building and paving purposes. Both the shales, and bluestones change westward into shales, and eastward become very coarse. In the Normanskill beds pebble layers alternate with the sandstones, while in both formations mud cracks are found in the shales and subsolifluction contortions in the sandstones, structures which show a slumping motion of the sands along the shore (Berckhemer, 21). The sandstones contain eurypterid and plant remains, the latter identified as Sphenophycus latifolius, and having a remarkably thick carbonaceous test which is so high in carbon that it will burn. In the shales occur the graptolites and eurypterids, the latter not being so abundant as in the sandstones, but exhibiting better preservation.
The sediments of both the Normanskill and the Schenectady were undoubtedly derived from the east as the following facts indicate: (1) Coarse materials, conglomerates and sandstones with intercalated shales in east along Schenectady-Catskill line, passing laterally into fine black shales westward in the Mohawk Valley; (2) deposits thicker and coarser in east than in west; (3) evidences of shore conditions in sun-cracks, wave marks, and subsolifluction, in east, of conditions in quieter water farther from shore in the fine black shales westward. Appalachia was the only large land area to the east from which siliceous sediments could come, and the characteristics of the sediments just noted clearly point to that continent as the source.
It will probably be readily accepted that the Normanskill and Schenectady are of terrigenous origin, especially since they are several thousand feet thick, but the point which is difficult of determination, is the origin of the fauna of these formations. That the sediments were fluviatile does not at all imply that the organisms in those sediments were also fluviatile. Indeed, it is usually argued that the presence of graptolites and "sea-weeds" in the same beds with the eurypterids is ample proof that all these types of life were marine and that they lived in the littoral zone in the sandy and muddy facies. First, in regard to the "sea-weed" Sphenophycus latifolius, there is no reason that I know of why such plant remains could not have been washed in from the land or might not have been living in the rivers, and have thus been swept into the sea. Secondly, it is evident that the presence of graptolites does not indicate deep sea conditions of quiet sedimentation as so often stated. Certainly, there is nothing incompatible with the assumption that the graptolites were spread out on mud flats, or river flood-plains as modern hydroids are, when washed in by the sea. At least the possibility must be granted that the pelagic graptolites would after death be more likely to float near or on the surface of the water until thoroughly decayed and disintegrated, rather than sink to the bottom, and be buried by sediments. In such a case, their only chance for preservation would be through stranding upon some surface where they could be quickly entombed by layers of mud or sand. This line of argument has only just been propounded by Professor Grabau, and while heretical it yet explains many curious occurrences.[3] Thus, although no definite statement can be made at present regarding the precise habitat of the graptolites, we may consider that it is reasonable to assume that their remains are chiefly found in formations accumulating near land especially on delta surfaces. If the graptolite-bearing beds are thin we may suppose that they were formed by frequent inundations from the sea, but when 1500 feet thick, as is the case of the Schenectady beds throughout the entire thickness of which graptolites occur at intervals, then the only interpretation is that the beds were a series of flood-plain and delta deposits, mostly above sea-level, and that the graptolites were stranded on the low-lying land areas by periodic incursions of the sea. It is often argued that such alternations of sandstones and shales as we see in this series indicate near-shore oscillatory conditions, the shales marking a slight advance of the sea and of fine deposition, the sands marking a retreat and the seaward advance of continental clastics. In the present instance it is difficult to explain the presence of eurypterids as marine organisms if we account for the lithological variation in the customary manner, for it is in the sandstones which mark the dominance of terrigenous sedimentation that the eurypterids are more abundant, while they are scarce in the shales which accompany the advance of the sea. If they were living along shore they should be abundant in the shales. The reply may be made that the eurypterids at that time preferred the sandy facies only and that the occurrence of dead individuals or shed exoskeletons in the muds was fortuitous. A phenomenon can hardly be called fortuitous which occurs again and again in response to a given set of conditions. Furthermore, if the eurypterids did live in the sandy facies then there is no reason why their remains should not have been preserved, for it is a mistake to believe that such exoskeletons would be destroyed by the waves, except perhaps on a shingly beach, a facies with which we are not here concerned. A short distance out to sea eurypterid remains would quickly be buried, the hollow case being soon filled by infiltrating sand. Anyone familiar with the occurrence of Limulus exoskeletons on sandy shores knows that they are easily filled by and buried in the sand and that they are preserved in toto, not broken to pieces. Thus we cannot account for the occurrences of the eurypterids on the assumption that they are marine organisms. In the Normanskill beds not a single entire specimen has been found, the whole fauna being made up mostly of carapaces with some separated abdominal segments. In the Schenectady sandstones the conditions are the same, but in the shales the preservation is better.
The evidence clearly militates against a marine habitat for the eurypterids in these two regions and the hypothesis of a fluviatile origin while not yet very strongly supported at least accounts for the observed facts. If the eurypterids were living in the rivers in Middle and Upper Ordovicic time, then it is to be expected that their remains would be carried out to sea. In rivers of moderate gradient it is not so likely that an abundance of remains of fluviatile organisms will be washed seaward, for they may be entirely broken up during transportation, or they may be caught in hollows along the banks, or even buried; but in eastern New York during the later stages of the Ordovicic the streams did not have a moderate gradient. The elevation of the land leading up to the tectonic movement known as the Taconic folding which displaced the rocks in some regions in the Hudson Valley as much as 90° was in progress at least throughout the Upper Ordovicic. One of the results of this movement was the steepening of the gradients of the streams which thus became torrential, not necessarily through increased rainfall, but through increased gradient. The streams consequently brought great quantities of clastic material to the margin of, and into the sea, where deposition probably went on in a sinking geosyncline.
Summary. The physical characters of the Normanskill and Schenectady beds point to Appalachia as the source of the sediments. The mode of occurrence of the eurypterids, graptolites and plant remains is better explained on the hypothesis that the eurypterids and perhaps the plants also were fluviatile and not marine organisms. As yet the Ordovicic merostome faunas are too little known to say that the habitat can be proved to be one or the other; the most that we can say is that all the known facts are better accounted for by the fluviatile hypothesis, which is fully supported by the palaeontological and chorological data.
3. THE SHAWANGUNK CONGLOMERATE[4]
In the intercalated shales in the Shawangunk conglomerate at Otisville, Orange County, New York, and at the Delaware Water Gap and elsewhere a large eurypterid fauna has been discovered. The Shawangunk is distributed in the form of a semi-cone, having its greatest thickness, about 2000 feet, in the Delaware Water Gap region, and thinning away in all directions. That is, it has the form of a dry delta or alluvial fan rather than of a sea coast deposit, for if it were the latter, it would be of fairly uniform thickness and would not have the semi-cone shape. The pebbles in the conglomeratic portion of the Shawangunk are well rounded, but in some sections a certain amount of angularity is still retained as a rule. For riverworn pebbles to be perfectly rounded, they must be transported for a considerable distance. Again the complete destruction of all but the quartz argues for prolonged transportation and frequent reworking. As in the case of the other clastics, there was no source for the conglomerate to the north, south or west, the only possible one being to the southeast where lay the old land of Appalachia. This is also indicated with great certainty by the shape of the cone of this formation which is thickest in the southeast, thinning out to the north, west and south. The material of this alluvial cone could only have been transported by rivers and its river-borne character and deposition upon land are well shown in the frequent occurrence of the torrential type of cross-bedding, and in the absence of any typical marine organisms, such as would be found in the subaqueous portion of a delta. Apparently there was no sea-border portion of this deposit, unless the Pittsford shale is considered as such. The Shawangunk conglomerates found in Pennsylvania and New York, thin away towards the region of Pittsford shale deposition, but actual connection has not been traced. There can, however, be little doubt that the Pittsford shale represents the finest material brought by the Shawangunk river from Appalachia. This mud was deposited very near the sea border, but there is no evidence that it was deposited in the sea, since the typical marine organisms are absent. It may be that the influx of fresh water was sufficient to keep these out. Interfingering with the shale, are the dolomite beds, deposited during short incursions of the sea (see section 4). If we could trace these Pittsford shales to the northern part of the state or into Canada, we would expect to find them grading into pure marine limestones, but no outcrops are accessible. The black shales intercalated between the conglomerate beds of the Shawangunk at Otisville and elsewhere represent the muds carried down by the river during times of flood. If the eurypterids were living in those rivers, their exoskeletons would have been floated down, while dead and even living individuals would have been swept down by the force of the torrential floods. The exoskeletons and floating bodies would settle down with the mud on the drying up of the water from the flooded areas. That such drying occurred is indicated by the presence of mud-cracks in these intercalated shales. When this mud dried up and became cracked, the eurypterid exoskeletons would be broken up and blown about by the wind, until the fragments should be covered by the next torrential deposit. This breaking up of the tests as the mud dried would account for the fact that the larger eurypterids are always found in a fragmentary condition, while the smaller ones are found whole. How to account for this fact on any other hypothesis is difficult to see.
4. THE PITTSFORD SHALE
This formation is typically developed in the western part of central New York where, in the town of Pittsford, Monroe county, the following section is shown (Sarle, 240, 1082):
| FEET | INCHES | ||
| 1. | Red shale | 6 | |
| 2. | Light gray, compact, fine-grained, dolomite, with imperfect conchoidal fracture, weathering light brown to cream color | 10 | |
| 3. | Soft, gritty mud-rock, purple with bright red mottlings | 1 | 3 |
| 4. | Dolomite like No. 2 | 4 | |
| 5. | Purple shale with red mottlings | 1 | 11 |
| 6. | Green shale | 1 | 2 |
| 7. | Thin layer dolomite like No. 2 | 4 | |
| 8. | Black shale, very compact, the base splitting unevenly; grading to olive-green shale in the upper part | 10 | |
| 9. | Dolomite like no. 2 | 10 | |
| 10. | Black shale, with leaf of dolomite inch thick four inches from its base | 1 | 2 |
| 11. | Dolomite like no. 2 | 2 | |
| 12. | Soft, green, arenaceous mud-rock, occasionally becoming shaly; the lowest exposed rock of the cut | 1 | 8 |
The eurypterid fauna occurs in the black shales, Nos. 8 and 10.
A more complete section is shown in the wells of the region, from which the exact location of the fossiliferous black shale beds is ascertainable. The section carries the series clown to the Lockport-Guelph horizon.
Salinan
| FEET | INCHES | ||
| 1. | Red shale or marlite | 10 | |
| 2. | Hard, fine grained, yellowish, dolomite, having an imperfect conchoidal fracture | 2 | |
| 3. | Red shale | 1 | |
| 4. | Break estimated at | 3 | |
| 5. | Dolomite like No. 2 | 3 | |
| 6. | Green shale or marlite | 4 | |
| 7. | Red shale | 1 | 8 |
| 8. | Break estimated at about | 2 | |
| 9. | Green shale | 2 | 5 |
Salinan—Continued
| FEET | INCHES | ||
| 10. | Black shale, very fine textured, fissile, and with 1 inch dolomite parting (eurypterid horizon) | 1 | 6 |
| 11. | Green shale | 1 | |
| 12. | Dolomite like No. 2 | 2 | |
| 13. | Green shale or marlite | 6 | |
From West Branch of Allen Creek
| 14. | Light colored waterlime, some pyrites and sun cracks | 5 | |
| 15. | Pea-green shaly marlite | 7 | |
| 51 | 7 |
Niagaran
| 16. | An impure yellowish prous limestone | ||
| 17. | Succeeded by an impure bituminous limestone made up of imbricating, shell-like domes, etc. |
It is thus seen that there are 41 feet 7 inches between the Lockport dolomites and the Vernon red shales, although there are two initial red beds, Nos. 7 and 3, showing that the red shale sedimentation was already in progress. A comparison of the two sections shows that the first bed of red shale (No. 7) comes at from 2 to 5 feet above the upper eurypterid-bearing bed, while the lower eurypterid bed (No. 10) occurs 21 feet above the typical Lockport-Guelph. In the lower interval are several thin beds of dolomite, and a waterlime showing sun-cracks occurs. The two shale beds are separated by a dolomite bed 10 inches thick, and in the lower black shale is a dolomite parting to 1 inch thick. The thin dolomite beds are often sun-cracked, indicating temporary exposure during formation.
The formations indicate a progressive change of conditions from those of Niagaran (Guelph) time when the widespread Stromatopora reefs were forming and the Guelph fauna flourished, through the period when impure dolomites were deposited in thin, ripple-marked layers containing some marine organisms and "fucoids," followed by conditions favorable to the formation of the impure bituminous limestone, to the final stage of the deposition of the impure porous limestone, 2 feet in thickness and containing a branching organism thought by Sarle to be a plant. Upon this bed lies the shaly marlite, the first of the lowest Salina or last of Upper Niagaran,[5] and above this a waterlime.
There is thus a marked change in the physical conditions which accompanied the withdrawal of the Niagaran sea and the initiation of the Salina type of deposits. In the lowest bed of the series occurs the last of the Niagaran fauna, Pterinea cf. emacerata. In the waterlime (no. 14) the same species occurs, together with a Lingula, Leperditia cf. scalaris and an Orthoceras. Then not again till the black shale of the eurypterid horizon, do these forms appear, but it is of great significance that these pure marine fossils do not occur in the eurypterid shale beds proper, but in the thin dolomite partings, and that in these partings the eurypterids are almost entirely wanting. The dolomites were evidently marine, but the conditions under which they were deposited "were not favorable to the eurypterids" says Sarle (240, 1086). The question immediately arises: why were the conditions not favorable? and then, where were the eurpterids during the intervals between shale deposition, i.e., during the time of marine dolomite deposition? If the eurypterids were inhabitants of the marine waters or of bays, estuaries, etc., they should be found in the beds containing the littoral fauna of the impure dolomite, for, as has been shown in Chapter III, the faunas of bays and other indentations along the shore are not restricted to such areas, but are much the same as the faunas extending all along a continental shore fine. Since the eurypterids were not living in the marine waters, where were they when there were no black muds forming? They appear suddenly in countless numbers representing six species, and they come with the black shale facies and disappear again just as suddenly. Their range, too, is very small. Sarle says: "Though the fine character of the silt forming the black shale and the evidence of interrupted sedimentation noted above, indicate slow accumulation, the occupation by the eurypterids was apparently of comparatively short duration, merely an incursion, as it were, since the black shale all told does not exceed 2 feet in thickness" (240, 1086).
To determine where the eurypterids were before their two sudden appearances, one must turn to the source of the black muds. These were not deposited in the open sea, for marine forms are wanting, and in any case, the mud must have been derived from the land. At that time the land to the west, north and south was all covered by the Niagaran limestones which would furnish pure clastic limestones and not impure siliceous muds. The only area from which the muds could be derived, was the land to the east. That the black mud was merely an extension of the muds forming at intervals on the Shawangunk delta must be obvious when it is seen that in that direction was the only source of the muds and that the Shawangunk muds contain the same eurypterid fauna. This will be more fully discussed in Chapter V.
The areal distribution of the Pittsford is limited. The shale is known from Monroe county and from Oneida county, New York. Both eastward and westward it dies out, the Vernon red shale resting directly upon the Niagaran. In a few localities black shales have been found which have been correlated with the Pittsford, but they contain no eurypterids. Such is the black shale at Buffalo, on Grand Island, and the dark shale in Herkimer county above the Lockport dolomite which contains no fauna except a few Lingulas. The outcrop in Oneida county is at Oriskany creek, where in a bluff occur some dark gray shales, about 21 feet below the base of the Vernon red shale, with intercalated waterlimes and dolomite beds. These dolomites contain fragments of one species, Eusarcus vaningeni, together with lingulas and orbiculoideas. Both the areal and vertical distribution, then, are limited, in much the same way as in the Bertie, and the source of this calcareous material may likewise be the same. (See beyond, p. 234.)
If the eurypterids of the Pittsford shale were brought in by the rivers coming from Appalachia, the waters in the region of deposition would become freshened by the inpouring of the river waters, and marine forms would thus be kept out. It is often assumed that the Pittsford shale marks a periodic increase in the salinity of the water, but in that case we are faced by a double problem: if the muds were not deposited by rivers, where did they come from, since they could not have originated in the sea? and then again, the question arises, where did the eurypterids suddenly come from? The muds might be æolian, but not the eurypterids. The only possible conclusion seems to be that the eurypterids and the black muds both were brought by rivers from the land, i.e., that the eurypterids were river-living organisms.
In this connection attention may be called to the fact that the species of eurypterids in the Pittsford and Shawangunk and to some extent the genera as well, are entirely different from those of the Bertie. This is not alone accounted for by difference in age, but is more especially due to difference in origin. The sediment of the Bertie and its fossils came from the continent of Atlantica, and those of the Pittsford from Appalachia. This is more fully discussed in a subsequent chapter (see p. 229).
5. THE BERTIE WATERLIME
The Bertie waterlime of Upper Siluric or Monroan age is confined to central and western New York, and the adjacent portion of Ontario, Canada. It is a gray, fine-grained, argillaceous calcilutyte of a remarkably uniform character, showing practically no variation in texture from place to place. Chemical analysis has shown it to be an impure limestone, high in magnesia, silica and alumina. The following analysis is that of an average specimen (39, 101).
| Si O₂ | 11.48 |
| Al₂O₃ | 17.50 |
| Iron | 0.90 |
| CaCO₃ | 42.75 |
| MgCO₃ | 20.35 |
| K₂O | 1.00 |
| Na Cl | 0.80 |
| Combined water and loss | 5.22 |
A typical section of the Bertie is exposed at Buffalo where Pohlman has recorded the following succession the lower part being obtained from borings. (See also Grabau, 82, 115).
Akron dolomite
| Feet | |||
| Bertie | Waterlime, about | 7 | |
| Shale and cement rock in thin streaks | 25 | ||
| Tolerably pure cement rock | 5 | ||
| Shale and cement rock in thin streaks | 13 | ||
| Camillus | Pure white gypsum | 4 | |
| Shale | 2 | ||
| White gypsum | 12 | ||
| Shale | 1 | ||
| White gypsum | 4 | ||
| Shale and gypsum, mottled | 7 | ||
| Drab colored shale with several thin layers of white gypsum | 58 | ||
| Dark colored limestone | 2 | ||
| Shale and limestone | 4 | ||
| Compact shale | 3 | ||
| Gypsum and shale, mottled and in streaks, approximately | 290 plus | ||
 Fig. 2. Sketch Map of New York Showing Location of Important Eurypterid-Bearing Beds 1, Buffalo and Williamsville; 2, Pittsford; 3, Waterville; 4, Litchfield and Cranes Corners; 5, Schenectady; 6, Otisville. |
In areal distribution the typical Bertie is not a continuous formation, but is found well developed at only two localities; namely, in Erie and in Herkimer Counties, New York, where the sediments were deposited in what Clarke and Ruedemann have called the Buffalo and Herkimer "pools." These two pools or basins are considered to have been of circumscribed area; the Buffalo pool extending from Bertie, Ontario, eastward into Erie County; the Herkimer pool being confined most of the time to the southern part of Herkimer County (See map, fig. 2). In spite of the faunules as a whole having such a restricted distribution, the Eurypterus lacustris of the Buffalo region has been found as far east as Union Springs, Cayuga County, although not at intermediate points, and E. remipes, the characteristic form of Herkimer County, has been found to the west at Waterville, town of Westmoreland, Oneida County, and still farther to the west in large numbers at Oriskany, Oneida County, at Cayuga Junction, Cayuga County, and possibly even at Buffalo. Dolichopterus macrochirus and Pterygotus cobbi are common to the two "pools."
Theories of Origin. A careful determination and a thorough understanding of the conditions under which the Bertie waterlime was deposited are essential in the attempt to determine the habitat of the organisms found in that rock. Because no one has yet given an exhaustive treatment of all possible conditions of deposition with a final singling out of the true one; and because, moreover, the answer to this question of deposition furnishes one of the most important lines of evidence concerning the habitat of the Eurypterida, I shall take up a detailed discussion of the subject. Such a fine-grained, stratified rock might have been deposited in one of the following four ways, and these appear to cover all possibilities: (a) by chemical precipitation; (b) by bacterial precipitation; (c) by the formation of an organic accumulation of calcareous shells or plants, or both; (d) by the accumulation of clastic or fragmental material.
(a) Chemical origin: That the Bertie waterlime could not have been deposited by chemical precipitation is amply shown by its stratification and especially by its composition. A rock which is a chemical precipitate, is more likely to be massive, never showing such fine stratification as is found in the Bertie, for in the process of chemical precipitation there is no arrangement of the material by currents bringing in fresh supplies which vary slightly in color or texture and which when deposited make the separate layers which produce stratification, since in precipitation the action is more or less continuous and minute crystals are formed which either entirely make up a rock, or else cement into a compact mass, fine particles of clastic material as is the case around modern coral reefs. The texture of a chemical precipitate would be a finely crystalline one, whereas the material of the Bertie does not conform to this, for a thin section of the water lime shows under the microscope an exceedingly fine-grained lime mud, the grains being angular and of varying sizes, with rhombic crystals of dolomite scattered through the mass of calcite fragments. There are also many fine, black specks, probably of carbonaceous material. The most significant fact of the composition, however, is the presence of the silica and the alumina, which forms nearly one-third of the rock. Such a composition is entirely incompatible with the idea of chemical deposition, where we should expect practically pure carbonates.
(b and c) Organic origin. If the Bertie were an organic deposit its fine texture would permit of only two types of organisms active in its formation, namely, the protozoa or the algae. The lime content might be supplied by Foraminifera or by lime-secreting algæ, the silica by Radiolaria. The microslide of the Bertie shows no trace of any of these organisms. One other method of organic deposition is possible. The work of Drew, Sanford, and Vaughan has recently shown that in warm or tropical seas certain bacteria are active in precipitating calcareous muds from the sea water. That the Bertie waterlime could not have had such an origin is evident from its chemical composition given on page 106 above, in which the silica and alumina play too important a part, amounting to 28.98 per cent of the whole.
Since the chemical and microscopic study of the Bertie proves the impossibility of either a chemical or an organic origin, we must conclude that the rock is clastic.
(d) Clastic origin. A rock of clastic origin may have one of two sources: (1') it may be composed of material which was originally derived from the sea, that is, it may be thalassigenous, or (2') it may be derived from the erosion or breaking up of a pre-existing rock on the land, that is, it may be of terrigenous origin.
(1') Organic material broken up in the sea by organisms, or along the shore by waves, consists of shells, corals, and other hard parts of organisms mixed with varying amounts of sands and muds, organic and inorganic, the composition depending on the character of the rock supplying the detritus. Such clastic deposits are especially well developed around coral reefs where the purely biogenic rocks grade laterally in all directions into the clastic ones. That the Bertie waterlime could not have been a lime mud derived from the erosion of coral or other reefs and deposited in the surrounding quieter water or in the lagoons, as in the case of the similar, fine-grained lime mud forming the Jurassic Plattenkalke of Solnhofen, is shown by the utter absence in this horizon or vicinity of reefs which could furnish such deposits, and again by the presence in the composition of the silica and alumina. In the Bertie the silica and the alumina is intimately mixed with the lime, as is shown by the relative constancy in composition and character of specimens from different parts of the formation. In the Plattenkalke of the Solnhofen, on the other hand, where the siliceous material represents the impure dust blown from the land, it is found in clayey layers (Fäulen) between the thin, bedded (Quicksteine) and thick bedded (Flinze) limestones, and not in intimate mixture with the other constituents, as is the case in the Bertie (293, 144, 209).
(2') The only remaining source of the deposit is the land, from which clastic material might be brought by the wind or by the rivers. If brought by the wind and deposited far enough from shore to be free from coarse material, the deposit would not have a circumscribed areal distribution. Such a restricted distribution is, however, possible if the material has been supplied by the rivers. If carried into the sea, it may be deposited in quiet water, and this may produce such a fine-grained rock as the waterlime, which is free from coarse clastics. Such regions of deposition would be found either far out at sea where all of the nearer-shore, coarser clastics were absent, or else near the shore, but in sheltered bays. If these river-borne muds were not carried into the sea, then they must have been deposited on land in the river flood-plains.
We may consider for a moment the possibility of this formation having been deposited at a sufficient distance from land to allow of the quiet accumulation of fine sediments, or else in sheltered areas along shore. Such deposits at the present time are represented by the blue or slate-colored muds, and these are the ones which are spread over the floors of shallow seas and out to the edge of the continental shelf. Murray and Renard (194) have estimated that these muds cover 14,500,000 square miles of the ocean floor. An average analysis shows the following composition:
| Ignition | 5.60 | CaCO₃ | 2.94 | |
| SiO₂ | 64.20 | Ca₃P₂O₈ | 1.39 | |
| Al₂O₃ | 13.55 | CaSO₄ | 0.42 | |
| Fe₂O₃ | 8.38 | MgCO₃ | 0.76 | |
| CaO | 2.51 | ——— | ||
| MgO | 0.25 | 100.00 |
A comparison of this analysis with that of the Bertie shows that the two types of deposits are as different as could well be imagined, the deep sea mud having combined alumina and silica 77.75 per cent, as opposed to 28.98 per cent, while the combined CaO and MgO is 5.00 per cent as compared to 63.10 per cent in the waterlime. One cannot argue much, however, from this pronounced difference between the two types, because it must be borne in mind that in the late Siluric the greater portion of exposed land areas in northern and western North America was covered with limestones or dolomites and that in consequence the muds which accumulated far out to sea, and which were the finest particles derived by the erosion of those land surfaces, would of necessity have been high in calcium and magnesium, whereas the blue muds accumulating in our present oceans are derived from a great diversity of rocks in which the limestones form a very small part. Thus, while we can find no analogous mud deposit in modern oceans, we are not justified in saying that such a one might not have formed in the past under different conditions; and I can, therefore, see no characteristics in the chemical composition of the rock to preclude the possibility of its deposition at a considerable distance from land. We are not, however, lacking in another criterion when the physical characteristics fail to be restrictive; the type of fauna represented is the safest guide in the interpretation of ancient regions of deposition. There is no region where muds are accumulating in the sea today, whether near shore or farther from land, where an abundance of organic remains is not being included. Along the entire Atlantic coast of North America the muddy facies of the littoral zone swarms with life, and while many of the species are confined to that facies it certainly cannot be claimed that where muds are accumulating there is a paucity of plant and animal life. Detailed studies of restricted areas of the ocean floor have proved that a large and varied fauna flourishes even where muds pour in in great quantities from the land. Thus, Walther (295, 36) has found that the muds in the Bay of Naples contain a fauna of about 1120 species of invertebrates and fishes. The fauna of the Bertie contains not two dozen species and nearly all of these belong to one phylum and to one class in that phylum, namely, the merostomes. Such a fauna cannot be considered as marine in any sense, if we accept the principles for the criteria of fossil faunas, based upon the study of recent faunas (p. 67 above). It is characteristic of no portion of the sea-shore, bays, lagoons, or estuaries, nor of the open sea, whether in the littoral belt or the deeper sea; such a fauna finds its counterpart in no waters of normal marine salinity, nor yet in those of modified marine salinity, either estuaries, epi-continental seas, lagoons, or other brackish to fresh water dependencies of the ocean. Thus, though we cannot determine with certainty the place of deposition of the muds from the chemical composition, or from other lithological characteristics, the fauna indicates with absolute certainty that those muds were not deposited in any portion of the sea.
From the foregoing discussion it appears that the Bertie waterlime is best interpreted as a deposit of clastic origin, and that the material was transported by rivers. It also appears that this material could not have been deposited in any part of the sea, for it has not the characters of non-terrigenous deep sea muds, nor the faunal content of a near shore, bay or estuarine deposit. There remains but one place for the deposition of these terrigenous muds and that is upon the land. There seems to be no escape from the conclusion that these lime muds of the Bertie represent the flood-plain or delta deposits from one or more rivers, or else that they accumulated as playa lake deposits. The characteristics of the sediments and faunas of such deposits have been fully described on pages 79-83, and it must be conceded that of all the known modes of deposition the lower flood-plain and upper delta regions of rivers come nearest in their physical and faunal characters to those found in the Bertie waterlime, though, of course, the nature of the sediment demands a source of supply in which calcareous material plays a dominant rôle.
It should be noted in this connection, that shallow water conditions of deposition for the waterlimes of New York and the associated calcilutytes (Manlius, etc.) are indicated by the occurrence of sun-cracked layers at several points. While these have not been found in the Bertie of the Buffalo region, they are wonderfully developed in the waterlimes of the Rosendale-Rondout regions, and in the Manlius of central New York and elsewhere.
Considering the waterlime as a flood-plain deposit, the history during Bertie time would be something like the following:
The early Siluric history of the eastern part of the North American continent had been admirably staged to lead up to the climax of waterlime deposition in many regions during the later Upper Siluric. During the Niagaran there had been a widespread advance of the sea which undoubtedly covered most of southeast and central Canada, as we may judge from the remnants still to be observed in the Lake Temiscaming region and elsewhere. At the base of the series is the Clinton followed chiefly by shales and limestones representing the Rochester and Lockport, and finally by a dolomite. Since the sea in which these deposits accumulated was a transgressing one, it is apparent that in some sections the Niagaran deposits would overlap the late Ordovicic deposits and come to rest directly upon the crystallines of the Canadian shield. Furthermore, progressively higher members of the Niagaran would come to rest upon the old land as the Niagaran sea continued to spread. By the end of Lockport time, the greatest expansion was reached, and contraction of the sea set in, the Guelph dolomites being deposited in this more circumscribed sea. In some sections the change in deposition is inaugurated by the argillaceous beds of the Eramosa formation, and some of the late Niagaran beds are somewhat argillaceous. Beyond the farthest line of expansion of the Niagaran sea, the crystallines continued to form the rocky surface of the land. The contraction of the sea continued, until by the beginning of Salina time it had shrunk to such an extent that only a small epi-continental sea remained. It makes little difference whether we assume that this sea dried up entirely during the period when the salt formed in central and western New York and in Michigan, or whether we believe that the contracted remnant of the Niagaran sea persisted, the greater part of the North American continent is known to have become dry land during Salina time. Many writers have pointed out the evidences of arid conditions in the Salina, and I need not here repeat them. The entire country was exposed to drying winds, rain fell but seldom, and then it came as cloudbursts, filling river channels quickly and creating torrential streams of short duration. Whatever vegetation there may have been upon that ancient land was destroyed by the heat, and we may picture the country as a great desert where desiccation was in progress and where the winds and the rivers of flood seasons were the chief agents of transportation for the mechanically broken up rocks. The Salina was by no means a period of short duration; the thickness of the salt deposits alone shows that a long time was required for their formation. Throughout this whole period, disintegration of the Niagaran and earlier limestones was in progress, until there must have been piled up great limestone and dolomite dunes with fine beds of impure clayey material wherever shales were exposed to the clastation processes of the semi-arid climate. The crystallines likewise suffered the same destruction, and they added their quota to the materials which were blown about in one of the earliest deserts recorded in the history of the rocks. This desert differed markedly from all the large ones which are known to us at present, in having a predominance of carbonates instead of silicates in the "sand" grains. We must not, however, push the doctrine of uniformitarianism too far and insist that all the deserts in the past must have been composed of siliceous grains, because that is the rule in modern large deserts. On a small scale limestone deserts are forming now, and if large areas of limestones could be exposed in the arid regions of Africa or Arabia these limestone deserts would form on a vast scale. But there is now too much diversity in the rocks of the earth's crust, because throughout most of the world the continents have in large part been above sea level during the Tertiary and Quaternary, and erosion has been going on so that many types of rocks are exposed and particularly large areas of crystallines, and when any or all of these are brought under arid climatic conditions, grains of a great range in composition are exposed to the sorting action of wind. In the Middle Siluric of North America, on the other hand, a land area which had been covered by limestone was subjected to arid conditions, and there is no escape from the fact that dominantly lime grains were formed by the prolonged exposure during which mechanical processes alone were active, and decomposition played no part.
Succeeding the arid or semi-arid climatic conditions of the Salina was a period of greater rainfall and of expansion of the epi-continental seas. The rivers became permanent in response to the rains of a pluvial climate, and there followed upon the period of rock destruction in situ a period of transportation of material from the land into the sea. The prolonged disintegration of the limestones and dolomites with local shales had provided a vast soil covering which must have extended to a considerable depth, and which, because of fineness and friability could easily be removed by streams. Even the weakest little rivulet would be able to carry a small load of this material, which was so conveniently prepared. With the increased moisture in the air decay became active in further breaking down the mechanically disintegrated rocks, and in this way the igneous rocks that were exposed through erosion would yield a certain amount of silica and alumina as would also the shale bands in the limestones. Thus, while the rivers carried material which was dominantly calcareous or magnesian, certain impurities were also included. Some difficulty has been offered by the high amount of alumina, to account for which I offer the following suggestion. The only decomposition product in which alumina is higher than silica is laterite which might have been formed either during the Salina, to the north of the desert in which the limestones were disintegrating, or else during the Monroan when the arkoses previously formed by mechanical breaking up were subjected to decomposition. That the northeastern portion of Atlantica was of a more pluvial character than the northern part which supplied the lime mud is independently inferred from the character of the deposits formed in western Europe at this time. For here the semi-arid conditions existed on the eastern side of the highland which supplied the sediment, indicating that the moist region lay to the west, where the great southward flowing rivers of Atlantica appear to have had their source.
So far it has been shown that the Bertie waterlime is of clastic origin, and that the sediments were river-transported from the north. The fine stratification of the deposit and layers of sun-cracks in certain localities are structural features indicating that the muds were deposited in quiet waters, while the nature of the fauna has shown that the place of deposition could not have been in the sea, either far from shore, or in any protected, littoral portion; the only remaining place is on the land. In concluding this discussion, therefore, we may test the hypothesis of the flood-plain or delta origin of the Bertie by determining whether it accounts for all the facts. We are to imagine, then, two rivers flowing from the low-lying Canadian area southward until they empty into the slowly-advancing Upper Siluric sea. Marine deposition would be active to the south and if the rocks now covering the Monroan in southern New York and northern Pennsylvania were removed, we would expect to find the mixed marine and freshwater beds which marked the interfingering of the delta deposit with those that were laid down in the sea. Unfortunately, at present we know only the marine Monroan limestones from Pennsylvania, the position of that ancient strand-line being nowhere exposed. If we bear in mind the fact that the outcrop of the Bertie waterlime in New York forms only a narrow belt extending east and west, it is readily understood that the cross-sections of the two eurypterid-bearing "pools" are to be interpreted as cross-sections of the two northsouth river channels (see figs. 3 and 4). The northward extension of those river courses has been removed by subsequent erosion, the southward continuation to the strand line is covered by later strata. If the Bertie waterlime of the two "pools" represents muds really deposited on flood-plains or the lower reaches of two rivers, then the lithological peculiarities of the deposit are readily explained. In that case we would expect these muds to become more marine southward, where they are now covered, and where the subaqueous delta part was situated. Between the deltas the Bertie should be impressed with certain marine characters, as it actually is in sections in Cayuga and Ontario Counties. In the Auburn-Geneva quadrangle, Cayuga
 Fig. 3. Block Diagram Illustrating the Two Principal Deltas of Bertie Time B, Buffalo; U. S., Union Springs, H, Herkimer. |
 Fig. 4. Cross Section (on the 43rd Parallel) of the Bertie and Herkimer Deltas |
County, the Bertie is an evenly bedded, impure, magnesian limestone, which when freshly broken is dark colored and of medium hardness. In portions it shows faint deposition lines, but heavier layers, from one to two feet thick, are usually quite compact. Some layers weather into a hard slaty shale. The fossils which have been found are: a few Lingulas, two species; one Orbiculoidea, a Rhynchonella, Leperditia alta, and fragments of eurypterids. In the Canandaigua-Naples quadrangle, Ontario County, the Bertie is a hard, dark, impure, hydraulic limestone, occurring in thick layers separated by thin seams of dark and apparently carbonaceous matter. The waterlime here shows a gradual transition from the Camillus. Fragments of eurypterid heads and appendages are not uncommon, and frequently Leperditia cf. alta, Whitfieldella laevis, and Leptostrophia varistriata occur. Yet the marine shells in both cases are seen to be of small specific gravity such as would easily be floated in across mud flats, and they evidently do not constitute a typical marine fauna since too few forms are represented. These occurrences of two or three species of brachiopods and of a crustacean in certain localities, far from proving that the Bertie as a whole was deposited in the littoral district of the sea, shows very clearly that the greater part of the waterlime was not deposited in any part of the sea and that only at intervals were a few marine organisms washed inland. Another significant fact that has already been referred to in connection with modern deposits is the separation of marine and fluviatile faunas in distinct layers. When river water meets with the invading tide, the current is checked and held back; this slack water is still fresh, and it deposits its load of mud and organic remains above the reach of marine waters. If marine currents later overcome the river currents and pass up the stream channel, marine organic remains may be deposited over the freshwater ones. Such lightweight structures as the exoskeletons of fluviatile crustacea and other arthropods are probably seldom carried out to sea against the opposing, denser salt water. If the eurypterids were fluviatile, the occurrence of their remains in abundance and well preserved in the regions where marine fossils are absent, and their scattered occurrence in the localities where a few brachiopods have been found is easily explained. Their entire absence from the Rosendale waterlime and the appearance of only a single specimen in the Rondout is likewise explained, since these deposits show a more marine character than does the Bertie of the Buffalo and Herkimer regions. The river portions of the Rondout and Rosendale either are not uncovered or else have been removed by erosion.
Summary. The only available source of the lime in the Bertie is from the muds derived by the erosion of an older magnesian limestone, the Niagaran, or in some cases, perhaps, the Trenton. Where the Bertie is eurypterid-bearing, the rock was evidently deposited above sea-level, as a river flood plain and subaerial delta deposit. Southward and laterally the subaqueous part of the delta carries few or no eurypterid remains, but more marine organisms. That the Bertie eurypterids lived in the rivers is thus indicated, while their absence from the Rosendale could be explained by assuming that the present exposures of these rocks are in the more marine portion of the deposit. The relations are shown in the following diagrams (figs. 5 and 6).
6. THE KOKOMO WATERLIME
The Kokomo waterlime of Indiana is of very much the same character as the Bertie waterlime, showing the same thin laminations and fine texture. Throughout a limestone series forty feet thick thin waterlime layers occur and it is in these alone that the films of eurypterid
 Fig. 5. Ideal N. W.-S. E. Cross Section from Buffalo, N. Y. to Tyrone, Pa., Showing Conditions During Bertie Time |
 Fig. 6. Generalized Cross-Section of the Same Region Showing Present Conditions due to Post-Bertie Deposition and Erosion |
exoskeletons are found. In the pure limestones a brachiopod fauna occurs, but no eurypterids are present; while in the separating waterlime eurypterids and ceratiocarids, but no brachiopods are found. Foerste has made the following statements in regard to the occurrence: "At the McReynold or Interurban quarry, in the southwestern corner of Kokomo, there is a much thicker exposure of the upper or brachiopod horizon. No merostomata have been found here.
"South of the center of Kokomo within the town limits, there is a deep quarry, covering a considerable area, where merostromata are common at an elevation of 3 to 3 feet above the base of the quarry. This belongs to the lower thinly laminated part of the section, and the richly fossiliferous brachiopod beds appear to be absent" (Foerste, 67, 7).
A section at the old George W. Defenbaugh quarry southeast of Kokomo, Indiana, shows the exact relation between the eurypterid-bearing layers and the brachiopod bed (67, 7).
| FEET | INCHES | |
| Heavy bedded fossiliferous limestone | 1 | 8 |
| Chert, thin, bedded, with ostracods | 1 | |
| Thin bedded fossiliferous limestone | 2 | |
| Base of brachiopod horizon | ||
| Darker layer of limestone | 2 | |
| Thin bedded limestone | 10 | |
| Heavier bedded limestone, but thinly laminated | 1 | 4 |
| Thin bedded limestone | 9 | |
| Darker limestone | 3 | |
| Layer with merostomata |
The line of reasoning which was adopted to show that the Bertie was a clastic, river-borne deposit which was spread out on the land
 Fig. 7. Section Southeast of Kokomo, Indiana, Showing Distinctness of Brachiopod and Eurypterid Horizons |
The second difference between the Kokomo and Bertie waterlime is that of origin, for while the latter was derived from the north the former must have come from the west since the sea covered the Michigan area during Monroe time and precluded the derivation of sediments from the Canadian region. It is difficult to arrive at an explanation of the lithogenesis of such a formation when so few outcrops are visible, but yet we can determine enough to show that the Kokomo sediment was river-borne and came from a continent to the west (see map, fig. 8). A study of the faunas convincingly shows the distinctness of the source of the material and organisms found at Kokomo (see below, pp. 253-256).
7. THE TARANNON-WENLOCK BEDS OF SOUTHERN SCOTLAND
Distribution of Formations. The clearest conception of the lithogenesis of the eurypterid-bearing Wenlock beds of southern Scotland is to be obtained from a survey of the palaeographic conditions existing in Great Britain from the end of Ordovicic time on through the Siluric. The outcrops in Wales, in the hilly areas of Cumberland and in innumerable outliers in Westmoreland and elsewhere, as well as those of the southern uplands of Scotland, indicate that throughout the Ordovicic the sea covered Wales, the greater part of western and central England and southern Scotland as far north as the great northeast-southwest fault line delimiting the northern edge of the tableland. The central and northern portions of Scotland formed a part of the old land which, rising to the east in the Scandinavian shield, extended westward through North Britain and Ireland on into the northern Atlantic, and which throughout the Palaeozoic furnished the sediments which were deposited either in the sea to the south of that ancient shoreline, or on the land to the north of
 Fig. 8. Palaeogeographic Map of North America During Bertie Time |
A rapid survey of the succession of events during Ordovicic time shows that there was a gradual retreat of the sea towards the south and southeast during the middle and upper Ordovicic and the lower Siluric, followed by a widespread advance during Wenlock time. A few of the typical sections will readily bring out these facts (see also the general description of the region on p. 151).
The Ordovicic and Siluric rocks of the Southern Uplands of Scotland are exposed in a series of belts trending northeast-southwest. The southernmost is a rather narrow, discontinuous strip composed of Wenlock and Ludlow flaggy grits and mudstones, bordering the northern coast of Solway Firth and extending northeast into the Cheviot Hills. The second belt, from 20 to 25 miles wide extends from St. Abbs Head on the east coast, through the Lammermuir Hills across the greater part of Selkirk, Peebles, Dumfries, Kirkcudbright and Wigtown (see map). This band consists of the Lower Siluric Llandovery and Tarannon beds. The third belt, narrow in the east where it does not quite reach the coast, but constituting the northern slopes of the Lammermuir Hills, broadens westward until it becomes 15 or more miles wide. It consists of Llandeilo and Caradoc limestones with a large amount of radiolarian chert of Arenig (Lower Ordovicic) age. The northwestern termination of this belt is the Girvan area with its great development of Arenig volcanic rocks. From 5 to 10 miles north of the third belt are two important regions one in the Pentland Hills, Edinburghshire, the other in Lanarkshire, where the Wenlock, Ludlow and Downton beds are exposed as inliers in the Old Red sandstone. The relation of these isolated Siluric outcrops to those of the southern tableland will be made clear by a consideration of the tectonic arrangement.
Towards the close of the Lanarkian a pronounced uplift took place accompanied by a tremendous amount of lateral compression giving a great series of folds whose axes run northeast-southwest, parallel to the major axis of the tableland. Denudation set in before the beginning of Old Red deposition so that the Old Red rests unconformally upon Siluric or Ordovicic beds. Moreover, formations which were continuous at the time of deposition now appear in far separated localities. Over the whole of this much folded and faulted series the Old Red sandstones were deposited by the rivers flowing south from the northern Highlands. Subsequent erosion has carried away large portions of these Devonic beds, and has cut down into even the lower rocks, so that the Ordovicic and Siluric are exposed in broad belts as shown above, while in certain places only inliers in the Old Red have as yet been exposed. To this class belong the isolated outcrops in the
 Fig. 9. Sketch Map of Southern Scotland Indicating Localities for Ordovicic and Siluric Eurypterid-Bearing Horizons |
Pentland Hills and in Lanarkshire. It is thought that in all probability the Wenlock and Ludlow in those regions were continuous and extended southwest into Ayrshire and northeast into the Lothians.
The Llandovery-Tarannon. In the lowest Ordovicic, volcanic activities were pronounced in the Girvan area, but throughout the central and northern belts of the tableland open marine conditions prevailed, marked either by submarine volcanoes or by the accumulation of radiolarian ooze, but the presence of fossiliferous mudstones in the northern belt of Arenig rocks indicates that the shore was not far distant. There is evidence that during Llandeilo time conditions were less stable in the northern area, for the black graptolite-bearing Glenkiln shales (Upper Llandeilo) often merge laterally into greywackes and grits, while sometimes, as for instance in sections at the headwaters of the Girvan River, the Glenkiln fossils occur in "minute dark seams in sandy shales, embedded in massive greywackes and grits" (Peach and Horne 215).
The section which is most complete, showing no disconformities and indicating, therefore, continuous deposition, is that at Moffatdale about 10 miles to the northeast from Moffat, where in the famous
 Fig. 10. Columnar Sections of Ordovicic in Moffat District, Scotland |
Dobb's Linn anticline studied by Lapworth, the succesions given in the first column, figure 10, is shown. It will be seen that the Glenkiln and Hartfell groups (Llandeilan and Caradocian, respectively), are complete and that the latter is followed by the Birkhill shales (Llandovery) which end with the Rastrites maximus zone (b3), which in turn is conformably followed by the green and grey shales of the Lower Tarannon. Crossing the strike to the northwest for about five miles, the Hartfell section is met with. It is the type locality for the shales of that name. The succeeding Birkhill shales are found to go no higher than the Monograptus gregarius zone (a3), which is conformably followed by the Tarannon grits. The significance of this succession will be spoken of presently. Continuing at right angles to the strike, there is found about 2 miles northwest of Hartfell on the Cow Linn, the last outcrop of the Monograptus gregarius zone. Only 3 miles northwest of this locality, near the junction of the Fruid water with the River Tweed the gregarius zone is no longer to be found, the highest of the Birkhill beds being the Diplograptus vesiculosus zone (a2) which is the second in the Lower Birkhill series. This zone is immediately followed by the Tarannon grits. As the last of the Llandovery outcrops are traced towards the north, fossils become very rare indeed and, although towards the boundary line of the northern and central belts no specimens of D. vesiculosus or of D. acuminatus have been found, a few other graptolites which along the valley of the Tweed are associated with these zonal fossils, have been encountered. It is thus seen that within the remarkably short distance of 9 miles,[6] as traced from the Dobb's Linn anticline to the Llandovery-Tarannon border, the whole of the Upper and nearly all of the Lower Birkhill shales have disappeared, the fossils becoming rare even in the shale members which,are found, and, most significant of all, the Tarannon grits or conglomerates everywhere follow upon whatever member of the Birkhill group forms the top of the section.
Such a stratigraphic relation might be interpreted in one of two ways. On the one hand it might be supposed that the Llandovery sea retreated to the southwest and that dry land conditions accompanied by subaerial denudation obtained in the areas laid bare. This would imply that more of the Birkhill shales had been deposited to the northwest of Moffatdale than are now seen in the sections and that the present exposures represent merely the parts which have not been touched by erosion. The Tarannon would then represent the river deposits spread out upon the eroded remnants of the Llandovery. That such is in all probability not the case is indicated by the statement made by Peach and Horne that in the Hartfell section (in the Frizzle Burn) "the black shales and mudstones of the Monograptus gregarius zone pass conformably upwards into the massive grits of Tarannon age without any representative of the Upper Birkhill Shales" (P. and H. 215, 133). The significant word is conformable. If the contact is conformable there was no erosion, and therefore it is not likely that two miles distant there was any considerable erosion. Thus another interpretation is called for. The facts, and they have been gleaned from a detailed study of many more sections than can be mentioned here, lead to the conclusion that there was no erosion between the deposition of the last of the Llandovery (Birkhill) beds, whether they were Lower or Upper Birkhill, and the lowest Tarannon beds. The structural relation is, therefore, one of replacing overlap, the Tarannon beds pushing to the southeast just as rapidly as the graptolite-bearing Llandovery muds retreated in the same direction. Lines of sections at right angles to the strike of the Llandovery and Tarannon rocks taken in various places from coast to coast, indicate that in the northern part of the central belt the Tarannon is always of a massive unfossiliferous character, grading southeastwards into graptolite-bearing shales and mudstones. There is no doubt that the coarse conglomerates and grits were river-borne. In one of the typical localities in the Moffat district Peach and Horne, in describing the conglomerate say, "the rock possesses a greywacke matrix, in which are embedded rounded pebbles of quartz, red chert with radiolaria, Arenig volcanic rocks, with boulders of granite and quartzite from eight to ten inches in diameter. Some small pieces of micaschist have also been observed. The fragments of quartzite and micaschist resemble rocks of those types in the Eastern Highlands; there can be little doubt that they were derived from that region" (P. and H. 215, 210). These authors also note in regard to the greywackes and grits that, "both volcanic and plutonic rocks have contributed to their formation. The fragments are angular or subangular. Well-rounded grains are rare. There is, further, a very great variability in the sizes of the constituent grains; indeed, the material does not appear to have been well sorted by aqueous action" (215, 211). It seems surprising that materials which had been transported by rivers for so great a distance, it being about one hundred miles from the Eastern Highlands to the Moffat district, should not be better sorted. However, it is clear that the material must have been brought down by rivers. That it was deposited as a subaerial delta or series of deltas which spread out into the sea to the southwest is suggested by the character of the materials, for in the extreme north of the Tarannon area there are only the coarse conglomerates without fossils, but these deposits merge ever so slowly into finer ones southward, the first change in the conditions of sedimentation being indicated by the intercalation of thin, leaf-like beds of shales bearing graptolites. Indeed, this domination of terrestrial over marine sediments is seen even towards the close of the Llandovery in many regions. In the northeastern portion of the Central Belt in the basin of the Gala Water, the various Birkhill zones are separated from each other by thick beds of grits, conglomerates, and greywackes. Even the graptolites show the effects of the great inpouring of fresh water, for not only are they rare, but those which are found are dwarfed as would be expected of a fauna dwelling in brackish water. Such features point beyond a doubt to the oscillatory conditions which prevailed along the shoreline and just so far as those conditions can be traced southward so far may we say the sea retreated in Llandovery and Tarannon time. It is only along the south central portion of the southern margin of the Central Belt that the highest Tarannon rocks are found; their continental origin is undoubted. They are unfossiliferous except for tracks and trails and they consist of grey, green, and red shales with bands of conglomerates one or two feet thick. All of these facts indicate a lithological replacement of marine by terrestrial deposits along a northwest-southeast line. The faunal replacement is equally striking. Along the northern border of the Tarannon belt the coarse deposits contain no fossils, but tracks and trails; when a few dark shale bands appear, they usually contain not good zonal fossils but a mixed Llandovery and Tarannon fauna. Monograptus exiguus is recognized as the lowest graptolite in the Tarannon and yet it frequently occurs with Rastrites maximus and Climacograptus normalis, the former of which is the zonal fossil for the uppermost Birkhill, and both of which are typical Llandovery forms. Towards the south, however, this interfingering and mingling of faunas is no longer noticeable and the Tarannon passes into the shaly, mudstone phase where zonal graptolites are well recognized, though in the passage to the upper Tarannon the mud facies is again replaced by conglomerates. The evidence supplied by the lithological and faunal characteristics, each taken independently, points conclusively to a replacing overlap and to the terrestrial origin of the Tarannon. The facts may be set forth in a generalized section.
 Fig. 11. Ideal Section Showing Restoration of Conditions During Llandovery-Tarannon Time in South Scotland |
The terrestrial origin of the Tarannon has been shown by two different and mutually independent lines of evidence. It is of interest, then, to find in the Upper Tarannon the fragment of a eurypterid. Near the southern border of the Central Belt just south of Bowden which is northeast of Selkirk, there is recorded the occurrence, in the grey blue shales and flagstones probably of the Hawick series, of the telson of Eurypterus and a fragment of Dictyocaris associated with crinoid stems. The typical Hawick rocks found in the neighborhood of Selkirk and further south at Hawick are themselves barren of all fossils but trails, burrows, and tracks. Near Selkirk Crossopodia and Myrionites have been found, while at Hawick Protovirgularia, Crossopodia, Menertites, Nereites and other tracks are abundant and the body segments of a Ceratiocaris have been found. The occurrence of the single eurypterid fragment in this great barren series is difficult to explain as a marine organism. It may be argued that the presence of crinoid stems is clear enough evidence of the marine nature of the deposits, but such disjointed stems might be washed out from an earlier deposit or even if of the same age as the eurypterid it is well known that those joints are swept great distances from the original habitat of the crinoids, and that they might be washed far inland on low-lying flats along the shore. At any rate, the single eurypterid remain fails to prove anything definite; it might be washed in from the sea, but then one must ask why the eurypterids are not found in the Tarannon muds in the regions where abundant graptolite faunas have been found. The fluviatile origin of the Tarannon has been amply shown, and it is easy to understand on the supposition that the eurypterids were living in the rivers, that fragments of the exoskeletons should be washed out from time to time. It may here be suggested that a further careful search in the Tarannon rocks might well yield a eurypterid fauna as fine and as unexpected as the fauna in the Shawangunk conglomerate. It is also not improbable that some of the tracks reported from the Hawick rocks were made by eurypterids, an interpretation in keeping with Patten's suggestion for the origin of Climatichnites (Patten, 206).
Following the retreatal phase of the Llandovery and the succeeding terrestrial phase of the Tarannon are the Wenlock beds. Though now exposed only in the southern belt below the tableland, in the small inliers in the Pentland Hills and Lanarkshire, and in the Girvan area, there is little doubt that the Wenlock at the time of its deposition extended entirely across this area. Such being the case, it represents the deposits of the advancing Wenlock sea. Continuous sections from the Tarannon into the Wenlock at various places in the belt south of the tableland show that the succession is conformable, thus proving that the line marking the end of the retreat of the sea must be northwest of this band and would lie, therefore, in the region of the tableland from which all Wenlock strata have, unfortunately, been removed. It is probable that the Tarannon-Wenlock shoreline was in the central or southern portion of the central belt. One of the finest exposures of the conformable contact of the Wenlock on the Tarannon is at Burrow Head, the outermost extremity of land between Luce and Wigtown bays just north of Solway Firth. The nature of the sediments along the belt south of the tableland indicates that oscillatory conditions prevailed, the seafloor being covered at times with fine muds, at others by coarse conglomerates. One would have expected that terrigenous deposits would have played a less important part there than in the Pentland Hills which are known to have been nearer the old shoreline. It is not unlikely that the land may have projected southward in a peninsula which lay between the present sites of Lanarkshire and Girvan, thus bringing the terrigenous sediments further south. In the southern belt there has been recorded a single occurrence of a eurypterid remain, so incomplete and so poorly preserved that it is specifically unidentifiable. Four miles south of Hawick at the junction of a small tributary with the Slitrig water Eurypterus sp. is reported associated with Ceratiocaris papilio and a number of graptolites. This type of occurrence, namely of a single eurypterid fragment associated with well-preserved abundant remains of marine organisms, has already been mentioned several times, and its significance pointed out. A general summary will be found below on page 194, in which the argument for the marine habitat of the eurypterids based upon such evidence is dealt with and, I trust, demolished for all time.
From the point of view of the determination of the habitat, we come now to the most significant occurrence of eurypterids thus far known in the British Isles. To be sure the "seraphims" of the Old Red Sandstone discovered by the stone-cutter of Cromarty and proclaimed by Agassiz to be "the remains of a huge lobster," are deservedly famous. Their size, abundance, association with the monster cephalaspid fishes, and above all the mystery attending their place in nature have shed upon the eurypterids of the Old Red Sandstone a picturesque and historical glow which makes the later discoveries of faunas merely so many cold triumphs of science. But the light which the Devonic merostomes throw upon the solution of the problem of the habitat cannot compare with that which emanates from the fauna of the Wenlock. And the reason is this: A large number of geologists have already come to the conclusion that the Old Red Sandstone was a series of torrential and flood plain deposits, in which case they can hardly fail to believe that the organisms found in the deposits were river-dwellers. Furthermore, it will not be a very difficult undertaking to convert the unbelievers in the river origin of the Old Red to staunch advocates of it. In fact, we may say that the case is so clear not only as to the lithogenesis of the deposits of the Old Red Sandstone, but also as to the medium in which the organisms of that time must have lived, that a few years' from now there will probably not be any thoughtful geologist who will not agree that the Devonic rivers supplied the sediments and were also the home of the Old Red fishes and merostomes. But in the case of the Wenlock it seems the rankest heresy to say that any of the organisms whose remains are found therein could be other than dwellers in the sea. The majority of palaeontologists would describe the Wenlock as exposed in the inliers north of the tableland in some such manner, "The Wenlock consists of a series of conglomerates, mudstones and grits with intercalated shale bands which are usually highly fossiliferous. While the coral fauna so characteristic of the Wenlock of England is lacking, there abounds, nevertheless, a representative marine assemblage which includes graptolites, brachiopods, pelecypods, gastropods, cephalopods, crustacea, and eurypterids. The merostome fauna is one of the largest known from a single horizon, comprising sixteen species, distributed in five genera, while the remains are so abundant that certain layers are almost like coal beds, they are so charged with carbon." Who, indeed, would have the temerity to claim that such a fauna of eurypterids with such associates could lead to any other conclusion but that the eurypterids dwelt in the Wenlock sea? It is just because such a conclusion in reality is entirely unjustifiable that I was led to state at the beginning of this paragraph that the eurypterid occurrence in the Wenlock is the most significant one in the British Isles when its interpretational value is taken into account.
Wenlock of the Pentland Hills. The Pentland Hills are formed from a series of the folded pre-Devonic beds and are completely surrounded by the various sub-divisions of the Old Red sandstone and by the igneous rocks. The Siluric rocks are exposed in four small isolated patches in these hills, and yet in spite of the small size of the outcrops and their isolation they have yielded more species of eurypterids than any other single formation in the world, with the exception of the Bertie waterlime, though there is this marked difference between the two faunas: whereas the Bertie fauna contains the most perfectly preserved individuals that have yet been recorded from any locality, with the exception of those from Oesel, the Pentland Hills fauna, on the other hand, is made up, for the most part, of fragmentary individuals.
The most important of these four inliers is that extending from the head of Lyne water to the head of North Esk River, a distance of about three miles. Although this inlier is the largest of the four it covers an area of only about two to three square miles. A number of excellent sections have been opened up by the North Esk and its various tributaries. The river itself cuts across the outcrop nearly at right angles, and since the beds here as in the other inliers are strongly folded, standing on end with the strike northeast-southwest, a considerable range in age is shown in the section, the lower beds appearing to the east, in the North Esk section where the Wenlock, Ludlow and a portion of the Lanarkian (Downtonian) are exposed, while to the west the Lyne water cuts through the passage beds or Downtonian. Of the numerous sections thus exposed the one which has now become most famous on account of the large eurypterid fauna discovered there by Hardie and Malcolm Laurie is that on the Gutterford Burn, a small tributary of the North Esk. (See map, fig. 12.) On the east bank of the burn, about a half a mile up from the North Esk Reservoir the strata consist of "flaggy micaceous greywackes" dipping at about 80 degrees to the southeast. Peach and Horne give a list of the fossils which they state come "from this band" (215, 593), but one may question the accuracy of this statement when
 Fig. 12. Sketch Map Giving Outcrops of the Wenlock of the Pentland Hills, Scotland |
compared with that made by Laurie at the time when he described the eurypterid fauna from these beds. He says: "The rock in which the Eurypterids are preserved is an irregularly fissile fine-grained sandstone, containing a considerable amount of carbonaceous matter distributed in thin layers. The only other recognizable fossil which occurs in the rock is the so-called Dictyocaris ramsayi, which occurs in considerable abundance" (145, 151). If one searches in the literature back to the time when eurypterid remains were first found in the Pentland Hills, one comes upon a description which is in but poor accord with the most recent one, emanating from the Scottish Survey, although bearing out quite well the statement made by Laurie. John Henderson in 1880 read a short account before the Geological Society of Edinburgh of some fossils which he had discovered in the Pentland Hills, in the beds in the Gutterford Burn. A few extracts from his paper will serve to give the clearest description which I have yet found of the eurypterid-bearing band. "This bed, which is upwards of a foot in thickness, is mostly made up of what I consider to be a mass of vegetable matter, along with an organism which has been described and figured by Mr. Salter . . . . as a large phyllopod crustacean under the name of Dictyocaris ramsayi . . . . there is in this bed a large amount of vegetable matter, some of the plant remains showing about one-tenth of an inch of carbon on the surface, and these plant remains are so associated with the supposed crustacean remains that it is difficult to determine the one from the other. I am now inclined to believe that the Dictyocaris is of vegetable origin. The fact of finding plant remains in such abundance in the Silurian rocks is, as far as I am aware, a new feature, as until lately true plant remains in that formation were considered doubtful. But there can be no mistaking their character and abundance in this bed, which is so thickly charged with carbon that it looks like an impure bed of coal . . . . the remains of undoubtful crustaceans of the genera Eurypterus and Stylonurus in a fair state of preservation occur in the same bed with the Dictyocaris and plant remains . . . . the bed in which these specimens are got must be at least as low as the Wenlock Shale. It lies several hundred feet below the fossiliferous bed in the North Esk above the reservoir, which is of undoubted Wenlock age" (115).
We do not know precisely the exact relation between the eurypterid-bearing bands and the beds containing the other fossils, but all available evidence indicates that they are not identical. Considering the decidedly different facies associated with the eurypterids and with the remaining fossils it seems probable that the former are confined to certain bands or lenses, as is often stated. In any case their occurrence is still capable of easy explanation whether they are actually in the same beds and really closely associated or whether they are found only along certain partings as is indeed indicated by Peach and Horne who say that "A characteristic feature of this eurypterid bed is the abundance on the divisional planes of that enigmatical fossil Dictyocaris ramsayi, forming, indeed, large black patches about an inch and a half across." One fact at least is clear: The eurypterids occur in a very thin band and are found in abundance in one section only, while a few fragments are found in one or two other nearby localities. The eurypterids are confined within a few inches vertically, while laterally the remains have a very limited extent, disappearing within a few yards. The occurrence at Gutterford Burn is in bed A of Brown and Henderson which contains ten species. Two species have been recorded from bed H to the northwest in a section exposed by the Henshaw Burn, a tributary of the North Esk. Except for these two isolated occurrences no eurypterids are found in the other beds or even in the same beds as they are traced along the strike. The remains of sixteen species representing five genera appear suddenly in a band a few inches thick, without forerunners in the underlying beds and with not a single straggler in the immediately overlying beds.
The following species of merostomes have been obtained from Gutterford Burn according to the identifications made by Malcolm Laurie (Peach and Horne, 215, 132, 593, 594):
- Bembicosoma pomphicus Laurie
- Stylonurus (Drepanopterus) pentlandicus (Laurie) emend. Clarke and R.
- S. (Drepanopterus) bembicoides (Laurie) emend. Clarke and Ruedemann
- S. (Drepanopterus) lobatus (Laurie) emend. Clarke and Ruedemann
- Eurypterus conicus Laurie
- E. minor Laurie
- Eusarcus scoticus (Laurie) emend. Clarke and Ruedemann
- Eurypterus 3 sp. undet.
- Stylonurus elegans Laurie
- S. macrophthalmus Laurie
- S. ornatus Laurie
- Slimonia dubia Laurie
- Dictyocaris ramsayi Salt
- Palaeophonus loudonensis Laurie
In beds of the same series in this section the following marine organisms have been obtained, according to Peach and Horne:
- Amphispongia sp.
- Nidulites favus Salt.
- Dictyonema venustum Lapw.
- Dictyonema (Chondrites) verisimile Salt.
- Cyrtograptus murchisoni ? Carr.
- Monograptus priodon Bronn.
- Monograptus vomerinus Nich.
- Favosites sp.
- Tentaculites tenuis Sow.
- Palasterina sp.
- Crinoid fragments.
- Lingula lewisi Sow.
- Lingula symondsi Salt.
- Strophomena walmstedti Lindst.
- Euomphalus rugosus Sow.
- Conularia monile Lindst.
- Conularia sowerbyi Def.
- Conularia sp.
- Orthoceras angulatum Wahl.
- Gomphoceras ellipticum ? M'Coy
The form described as Bembicosoma pomphicus is somewhat problematical and may be more nearly related to Hemiaspis than to the eurypterids. Of special interest is the finding of a scorpion in these beds. There is only one specimen of Palaeophonus loudenensis and this is in a bad state of preservation so that it is impossible to tell whether this early scorpion was aquatic and gill-bearing, or terrestrial. Drepanopterus is a new genus founded by Laurie for the three species: pentlandicus, bembicoides, and lobatus, but they are all described from imperfect material; the first from a fairly good specimen, the other two from a few slightly better.[7] Laurie's genus Bembicosoma[8] with the one species pomphicus was established for a few rather good fragments, which, however, were only doubtfully identified. Slimonia dubia, also described by Laurie, is but poorly represented, as are likewise the three species of Stylonurus: elegans, ornatus and macrophthalmus. Imperfect remains are all that have been found of the five species of Eurypterus: scoticus, punctatus, minor, cyclopthalmus and conicus.
In the Girvan area near Straiton where continuous marine, though near-shore deposition was going on through Tarannon on into Wenlock time, the strata are found to be highly fossiliferous at certain horizons and graptolite bands are well made out. Collections made in a quarry near Blair Farmhouse not far from the village of Straiton have yielded a number of fossils, among which is a Eurypterus sp. Owing to the fact that British geologists seldom state the exact horizon at which fossils are collected, and of course this is often difficult to do when the strata stand on end and break off in slabs from which collections are made, it is impossible to say whether the eurypterid occurred in a seam with the crustacea while the undoubted marine forms occurred in other seams, as is found so often to be the case. The list given by Peach and Horne is merely quoted as coming from this quarry (215, 549).
- Eurypterus sp.
- Beyrichia kloedeni (M'Coy)
- B. impendens (Jones)
- Entomis globulosa (Jones)
- Monograptus galaensis (Lapw.)
- M. priodon (Bronn)
- M. riccartonensis (Lapw.)
- M. vomerinus (Nich.)
- M. sp.
- Retiolites geinitzianus (Barr.)
- Favosites sp.
- Lingula symondsi (Salt)
- L. sp.
- Orthis sp.
- Siphonotreta anglica (Morris)
- Cardiola fibrosa (Sow.)
- Bellerophon sp.
- Orthoceras subundulatum (Portl.)
After this rather careful study of the occurrences in the Wenlock, we are in a position to form a valuation of the hypothetical description of this formation and its fauna, which I gave on page 130 above, and which may be fairly taken as the prevalent view expressed in textbooks and by authors generally. The main statement that is dwelt upon insistently, and which is so dangerously plausible, is that in the Wenlock the eurypterids are found in an undoubted marine formation in association with typical marine fossils. Such a statement is full not of truths, but of half truths, and these are far harder to combat than actual untruths. In the present instance we have only to consider how much weight would attach to such a statement if the following significant bits of information were added: (1) The eurypterids do not occur in the same beds with the undoubted marine forms, but always in certain leaf-thin bands which carry no other fossils except Dictyocaris ramsayi, a form which may be a fluviatile if not a terrestrial plant or animal, but the systematic position of which is at present wholly undetermined. The thickness of all the beds containing the eurypterid-bearing bands is only one foot; it is a grit and greywacke; the typical marine fossils are found in the shales and in limestone lenses. (2) Not a single complete eurypterid has been found among the hundreds of specimens collected and the very species described by Laurie have usually been founded upon fragments. (3) The exoskeletons have not only been dismembered—this might be expected in any case, for during the process of decay the various members might fall apart and thus be embedded in the mud, few complete individuals being entombed—but the fragments are macerated, the edges are broken and worn, the surface sculpturing indistinct, altogether showing evident signs of wear. (4) The occurrences are not widespread; although many good sections are obtainable in the Pentland Hills inlier, in only two places are eurypterids found. In both cases the remains are confined to bands a few inches thick, extending laterally but a few feet. (5) The eurypterid fauna appearssuddenly with no forerunners and no descendants so far as may be judged from the faunas of the beds immediately below and above the bands containing eurypterids; two of the five genera are confined to this horizon. (The full significance of statements (4) and (5) will be taken up in the next chapter on distribution.)
These are the facts which are generally not mentioned when the eurypterids are declared to be abundantly represented and associated with a good marine fauna. These five facts seem to be difficult of explanation if it be supposed that the eurypterids were marine organisms. How are we to account for the fact that the merostomes accompanied by that form which so often occurs with them, Dictyocaris ramsayi, are found in layers separated from those containing the brachiopods, pelecypods, etc.? If the eurypterids lived in the marine waters, why are their remains not found with those of the other marine organisms? Band D contains many genera, species, and individuals of brachiopods, pelecypods, gastropods, cephalopods, crustacea, crinoids, trilobites, corals, the first three groups being especially well represented, the last three by only two or three species. But not a trace of a eurypterid is found in this fauna which, according to the criteria given on p. 76 is a typical marine fauna. Moreover, why should their appearance be so sudden and so localized? The eurypterid band dies out within a few yards of Gutterford Burn, and in the other sections where one would expect to find eurypterids again, for the associated marine fossils occur, there is not even a fragment? I must confess that such anomalies in distribution are not compatible with a marine habitat for the organisms so affected. Again, why should the eurypterids have suffered such maceration, while the remains of the other organisms were entombed in a perfect state? Wonderfully preserved starfishes and trilobites are found in one of the beds, the marvellous brachiopod fauna of Band D has been described and figured by Davidson, and his book amply attests to the abundance and fine preservation of the molluscoidea; but the eurypterids are broken up, often unidentifiable and never what the palaeontologist would consider good material. These questions have been, or will be, fully treated in the appropriate sections on the bionomy and distribution of the eurypterid faunas. We are here concerned only with the lithogenesis of the beds so far as that may be separated from faunal considerations. From the study of the sections, and the lithological characters of the rocks, I offer the following interpretation for the eurypterid-bearing horizons of the Wenlock of southern Scotland.
It will be remembered that the Tarannon marked a period of retreat of the sea toward the south, the shoreline being at the end of that time somewhere in the central part of the Central Belt, while there must have been an embayment in the Girvan area where continuous marine sedimentation was active. The Wenlock sea which covered the greater part of western, central, and northern England at least, advanced over the terrestrial beds which were deposited during Tarannon time. Unfortunately, the most critical areas of deposition of the Wenlock have been removed by subsequent erosion. The Southern Belt was in the region of continuous marine deposition from Tarannon into Wenlock time, as the sections in many places show. Even in the Pentland Hills the base of the Wenlock is nowhere visible, the beds standing on end for the most part and sticking up through the Old Red Sandstone. In Lanarkshire, however, there occurs in the Lesmaghagow inlier only, below the Ludlow, a series of blue greywackes with shale partings which is 1300 feet thick and has proved unfossiliferous throughout except in one locality where a few specimens of Murchisonia (specifically undeterminable) and some doubtful forms called Orthis have been found. It has been thought by the Scottish Survey that part at least of this formation was Wenlock in age. I should like to offer the following purely theoretical suggestion. It has been shown that during Tarannon time rivers flowing from the Eastern Highlands carried down pebbles and boulders which were deposited in the Central Belt. Probably the whole of central and northern Scotland was above water then, and either subaerial erosion or deposition was going on. The 1300 feet of greywackes and shales below the Ludlow in Lanarkshire might represent in their lower part delta or torrential deposits accumulated during Tarannon time and in their later part similar deposits during Wenlock time. Their unfossiliferous character and great thickness would thus be accounted for. Future study in those rocks should be directed towards the search for cross-bedding, if any, and the type represented, for plant remains, tracks, and eurypterids. As the Wenlock sea advanced northwards—there is little reason to doubt that it did, for the same marine fossils are found in the Southern Belt and in the Pentland Hills—it reworked the Tarannon, and a basal sandstone, conglomerate or shale was formed, depending upon the nature of the Tarannon continent where the sea transgressed. Thus along the northern border of the Southern Belt the basal Wenlock consists of "greenish grey, flaggy grits, separated by grey shale bands, some of which are crowded with Crossopodia, Nemertites, and other tracks, resembling those found in the Hawick Rocks." On the Slitrig Water the shales and surfaces of the greywackes are crowded with tracks. These rocks pass conformably downwards into the Tarannon rocks of Hawick, indicating that the actual seashore at the close of Tarannon and beginning of Wenlock time must have been just about in this region. The unfossiliferous grits and greywackes (the first division of the Wenlock) appear along the northern border, while the second division occupies all of the rest of the belt to the south except for small patches or inliers in the extreme south where the third division of the Wenlock is seen; there are also inliers in various localities throughout the belt east of Langholm, showing the basal grits and greywackes projecting up through the second division of the Wenlock. This distribution indicates the presence of the basal sandstone of the advancing sea throughout the southern belt. The single eurypterid fragment found in this belt, it will be recalled, was discovered in the track-crowded shales and greywackes at Slitrig Water. These being interpreted as basal beds of an advancing sea, it is most natural to expect that the sea, rolling landwards and up the rivers, slowly but unceasingly converting the dry land into seafloor, should catch river-dwellers who were not able to or did not migrate upstream fast enough, and even if there were none such, at least dead remains would inevitably be passed over by the sea in its continued advance. One would undoubtedly expect more than a single fragment and probably more will be found in the southern Wenlock rocks. The more abundant occurrence in the Pentland Hills is explainable on the supposition that the sandy bands containing the broken exoskeletons represent the outwash from rivers into the sea, of shed exoskeletons and maybe even of the remains of eurypterids which were killed off in great numbers by the entrance of salt water into the rivers. So soon as this group of organisms was able to migrate far enough away from the sea which had overtaken the earlier individuals, the appearance of exoskeletons in that region would come to an end, but one would expect similar catastrophes to occur in another locality at a higher horizon. Unfortunately, the exact method of entombment must remain hypothetical, since the exposures are so few, but that the eurypterids did not live in the Wenlock sea is apparent. One further argument which might be adduced is that in the purely marine, open-sea Wenlock of England, not a trace of a eurypterid has been found, although if they were true marine organisms during Lower Siluric time as most geologists claim, then it is surprising that they alone of the marine fauna should be found only in southern Scotland although migration was open along most convenient marine channels into Wales.
8. UPPER SILURIC OF OESEL
For beauty and perfection of preservation no other known eurypterid remains can compare with those from the island of Oesel. Though only five species have been found and only one in abundance, the lack of a varied fauna is entirely compensated for by the rare condition of the fossils. After tens of millions of years the exoskeletons of these organisms now so long extinct appear in the rock, differing not in appearance from the shed skin of a Limulus buried in the sand today. We must be filled with awe and with the profoundest admiration for the marvellous ways of nature, when we look upon these remains unchanged in chemical or physical characters during all the aeons which have passed since they were entombed, still retaining the brown color so familiar in modern horseshoe crabs, with the very chitin of the test unimpaired, while even the brittle exoskeleton itself, at times, can be removed from the rock intact.
 Fig. 13. Sketch Map of Oesel for Upper Siluric Localities |
History of Discoveries. This fauna was discovered in 1852 by Dr. Alexander Schrenk, during a trip made for the purpose of studying the Ordovicic and Siluric rocks of the northwest provinces of Russia, namely, Livland and Estland, and of the adjoining islands Oesel, Dago, Moon, Worms, etc. On the first and largest of these islands he found outcrops of the uppermost Siluric rocks in the town of Rootziküll (see map, fig. 13) and there he came upon the first of the eurypterid fauna which was to become world renowned.[9] In his report on this region he says: "The gray, compact dolomite of Rootziküll, on the west coast of Oesel, reveals the thin membraneous tests of Eurypterus remipes Dekay [= E. fischeri Eichwald] entirely unchanged, not only in their chemical composition, as pure chitin, like that found in the shells of living Crustacea, but also in their whole internal microscopic structure and preserved with their original brown color peculiar to living animals" (Schrenk, 254, 35).
In the following year, 1853, Eichwald apparently not knowing of Schrenk's discoveries visited the same provinces and islands and on Oesel two versts from Rootziküll in the village of Wita he, too, came upon the eurypterid horizon whose assemblage of organisms surprised him not a little, for he says: "I was astonished to find a vast multitude of Eurypterus remipes [E. fischeri (Eichwald)] in this limestone" (Eichwald, 57, 40). By his collections he added much to the knowledge of the rest of the fauna, but I shall not at this point give the species which he found, since later workers added materially to the faunal lists. During that same summer Schmidt and Harder accompanied Eichwald to Wita and other nearby localities where eurypterids were found; in 1856 Schmidt returned again to Oesel and the following year Niezkowski, Schmidt and Czekanowski made large collections at the best localities. Again in 1858 Schmidt revisited the island, and as the result of these extensive collections and field studies several important papers were brought out. By far the most complete and comprehensive were those by Schmidt, the first published in 1858 entitled "Untersuchungen ueber die Silurische Formation von Ehstland, Nord-Livland, and Oesel"[10] embodies the first detailed stratigraphic and palaeontologic discussion of these regions. Schmidt gave the first geologic map of the region and the zonal subdivision of the "Silurian" which is still used in the east Baltic provinces. In the following year Schmidt published a short notice on some further discoveries in Oesel (243). His most important paper on this island appeared a number of years later in 1883 as one of the "Miscellania Silurica" in the Memoires de l'Academie Imperiale des Sciences de Saint-Petersbourg, entitled "Die Crustaceenfauna der Eurypterenschichten von Rootziküll auf Oesel." Precise information is given regarding localities, species are fully described and compared with related forms and excellent illustrations are given, so that with these papers and one by Nieszkowski in 1859 on "Der Eurypterus remipes aus den obersilurischen Schichten der Insel Oesel" (197) one may gain an accurate knowledge of the fauna and the sediments. Notes by Nieszkowski in connection with his work on the trilobites have proved helpful, and for further details the reader is referred to the titles under his name in the bibliography as well as to numerous papers by Schmidt.
General Stratigraphy. The Siluric exposures on the island of Oesel include two divisions: the lower Oesel group or zone I, and the upper Oesel group or zone K of Schmidt. The strata have a gentle dip to the south so that higher and higher beds appear in that direction. The lower beds, of Wenlock age, cover a considerable part of the northern half of the island, while the upper or Ludlow beds are found in the southern portion (see map, fig. 13). In the extreme north the lowest part of zone I occurs carrying typical Wenlock fossils; southward, as on the peninsula of Taggamois the upper division of the zone is exposed, showing well its dolomitic reef structure; bryozoa, crinoids and brachiopods are abundant, and do not differ essentially from the forms in the underlying marls. The last exposures of the upper part of zone I yield abundant Thecia swindemana, a coral found in the Upper Visby beds of Gotland, also Leperditia baltica, which occurs in divisions V, VI and VII of North Gotland, Strophomena imbrex, found throughout the Wenlock or lower divisions in Gotland, and Zaphrentis conulus, characteristic of the upper part of the Visby formation (III) immediately below the Pterygotus marl of Gotland. This higher portion of zone I is to be correlated with the Leperditia baltica zone of Gotland (Schmidt, 250, 132).
Throughout the entire south and southwestern parts of the island, zone I is succeeded by zone K, but the actual contact is nowhere observable. This zone likewise shows two subdivisions, a lower, made up of thin-bedded "plattenkalk" or dolomite, in some places unfossiliferous, in others carrying eurypterids and fishes, and an upper very fossiliferous horizon known as the Ilionia beds on account of the abundance of that pelecypod. The Ilionia, beds are to be correlated with zone VI of Gotland which is of Upper Ludlow age. Some of the diagnostic Upper Ludlow fossils recorded from this horizon in Oesel are: Ilionia prisca, Megalomus gotlandicus, which occurs just above the Ilionia beds in Gotland, Murchisonia compressa (Gotland VI), and Spirigera (= Meristina) didyma, which is the most widespread form in the northern outcrops of zone K in Oesel and which occurs at Visby in the top of bed III, in the Sphærocodium marl below the Ilionia limestone, and above the eurypterid marl of Gotland, as well as in the Aymestry limestone and Dayia beds of England, all (except possibly the last two) of Upper Ludlow age. Thus there is evidence of a faunal break in the series, since beds containing Upper Ludlow fossils everywhere in eastwest sections across central Oesel follow upon beds with Wenlock fossils. In many localities the indications of a physical break are also present as may be best shown in a few detailed sections.
The fullest development of the eurypterid fauna is seen in the rocks underlying Rootziküll on the west coast of the island of Oesel in the parish of Kielkond. Here the beds of the lower part of zone K are a fine-grained "plattenkalk" or dolomitic calcilutyte, in which the chitinous exoskeletons of Eurypterus fischeri Eichwald, E. laticeps Schmidt and Pterygotus osiliensis Schmidt have been so excellently preserved. Associated with the eurypterids in the same bed have been found the tail of Ceratiocaris notlingi Schmidt,[11] the shields of two cephalaspid fishes Thyestes verrucosus Eichw. and Tremataspis schrenkii Schmidt, and the shells of the little Lingula nana Eichwald. Nearly fifty years after these first discoveries Schmidt was able to add a new species to the fauna perhaps representing a genus not heretofore known outside of North America. From A. Simonson he obtained a slab which showed the portion of the abdomen and carapace of this new species which he called Stylonurus (?) simonsoni (252, 157).
Attention has already been called to the fact that the eurypterid exoskeletons have the original chitin still preserved and that this may be lifted from the rock so that both the upper and under surface and the sculpture thereon may be studied. The shells of the remaining fossils which are found in this bed are destroyed; these include the rarely occurring Hemiaspids: Bunodes lunula Eichw., B. rugosus Nieszk. and B. schrenckii Nieszk sp. as well as Pseudoniscus aculeatus Nieszk. and the shells of Orthoceras tenue Eichw. All of these forms are represented only by carbonaceous films. In the environs of Rootziküll the eurypterid-bearing plattenkalk appears at the surface everywhere and as Schmidt puts it, "In the extent of a single verst one may here lay out places for eurypterid quarries to one's heart content" (248, 29).
Above the plattenkalk horizon is a brecciated limestone of no great thickness consisting of angular or slightly rounded fragments of compact limestone in a matrix of similar limestone which contains Calamopora polymorpha. The breccia is not derived from the underlying dolomite, according to Schrenk (254, 47). This physical evidence, of a break at the top of the eurypterid dolomite has been more fully described from other localities, as, for instance, at Wita, the section next to be considered.
To the southwest of Rootziküll is the village of Wita. Here in the yellowish white dolomite which is the characteristic eurypterid-bearing facies two quarries have been opened. It was found that the eurypterids occurred not only in the dolomite, but also at a higher horizon in a brecciated coral limestone which is made up of angular, sometimes rounded white nodular masses which are for the most part corals lying embedded in a uniform, yellow, marly limestone matrix. Schmidt (241, 167, 168) would correlate this bed with the Burgsvick oölite of Gotland, the formation which there marks the break between the upper and lower Gotlandian. The limestone at Wita is only one foot thick; in its upper part it contains Leperditia baltica, Turritella obsoleta (= Holopella obsoleta), Spirifer elevatus, and certain corals, all being characteristic of the Upper Ludlow of England and of the Upper Gotlandian of Gotland. In the lower portion of the breccia occur: Cephalaspis verrucosus, C. schrenkii, Eurypterus fischeri, Bunodes lunula, a new crustacean Dithyrocaris ? sp., Orthoceras bullatum ?, Lingula nana, and Palæophycus acicula, besides many fragments of crustacean claws, segments of walking legs and the like. The section is of importance for three reasons: (1) There is physical evidence of a break at the end of Wenlock or Lower Ludlow time, marking a retreat of the sea. It did not return until Upper Ludlow time as indicated by the presence of fossils of that age in the matrix of the brecciated limestone. (2) The eurypterids occur abundantly in the beds deposited immediately after the normal marine conditions ended, while the sea retreated, and at the time when dry land was being enlarged and consequently rivers were extending their distal portions. (3) The eurypterids are also sparingly found in the breccia and conglomerate which marks the return of the sea and renewed deposition of marine sediments with marine organic remains. (4) The eurypterids do not occur in the beds with the marine fossils but always in distinct zones a few inches thick, their whole representation being confined to not more than a few feet in the entire Oesel series.
West of Rootziküll a distance of about 5 versts there is an exposure not far from Gesinde Wessiko Maddis along a little brook which rises near Lümmada, but is usually dried up. Here the lower rock is limey, not dolomitic and the eurypterids are not very abundant, but the rock above is crowded with Platyschisma helicites, Leperditia phaseolus, and the delicate fish scales of Coelolepis schmidti Pander together with fragments of seventeen other species of fishes (Schmidt, 241, 168, 248, 29). The upper beds are evidently the continuation of the brecciated limestone of Wita. Proceeding in the same south-westerly direction from Rootziküll towards the coast one comes to the Attel estates or Gut Attel where there is a small outcrop of yellow, coralline limestone which on exposure weathers white and which carries Stromatopora sp., Cyathophyllum, Favosites hisingeri, and F. fibrosa; similar brecciated inclusions occur as at Wita. A little farther to the west in the village of Attel may be seen on the west side of the deeply indented bay a coarsely crystalline yellow-dolomite and beneath this is the coral limestone of the Attel estates which here is not entirely composed of corals, but contains also Eurypterus fischeri, Lepeditia baltica, Orthoceras bullatum, and Murchisonia cingulata = M. compressa, the last being the same species which is found in zone VI in Gotland. It is clear that the eurypterid remains at Attel are found not in the plattenkalk beds, which here are barren of all organic remains, but in the overlying coral limestones (Nieszkowski, 197, 307; Schmidt, 241, 169, 170). The last section in this series is at the Soegi-ninna point about 12 versts from Rootziküll, where the rock walls rise from the sea to a height of 10 or 12 feet. In the upper part is seen the typical crystalline dolomite with nodular inclusions which here and there give place to thin, unaltered limestone beds with Leperditia baltica and Murchisonia compressa; the lower part of the rock walls consists of platten dolomites which appear to be the continuation of those of Wita, but which have not yet yielded any eurypterid remains after fifty years of patient search (Schmidt, 151, 169, 170).
The outcrops in the southeastern portion of Oesel show only traces of eurypterids here and there. For instance, between Uddafer and Ladjal, north of Arensburg, Schmidt found in small ditches along the roadside Phragmoceras sp., Spirigerina prunum, S. didyma, Pleurorhynchus sp., Laceripora cribrosa, but no eurypterids. In the quarry at Ladjal itself, in a band of limestone apparently in place, there occurred a great mass of Leperditia baltica, and also Spirigerina didyma, while in marly interbedded layers Eurypterus fischeri occurred in traces. To the southeast this limestone merges into solid gray limestones carrying trilobites, crinoids, brachiopods, etc., but not eurypterids. At Nessoma, southeast of Sandel occurs an outcrop of the upper crystalline limestones which marks the Spirigerina prunum horizon, and in intercalated brown marly layers were found great numbers of fish scales and breast plates, similar to those occurring at Ohhessare-Pank on the southwestern end of the island. The section at Lode about the same distance west of Arensburg as Nessoma is east of it, has brought to light one of the richest collecting grounds on the island for the typical marine forms. Here the rock is a gray limestone in which Spirigerina prunum occurs in great numbers but is not well preserved; Leperditia baltica is occasionally found, but the abundant forms are: Calymene blumenbachii, Orthoceras bullatum, Spirifer elevatus, Orthis orbicularis and Chonetes striatella, all characteristic of the Upper Ludlow of England (Schmidt, 241, 176–7).
In summary, it may be said that the detailed sections bring out the sporadic occurrence of the eurypterids in very thin beds, rarely intimately associated with the typical marine forms which occur in beds above and below the eurypterid marls. As the beds are traced to the south, southwest and southeast they are seen to be replaced by those containing a pure and abundant marine fauna, but not a trace of a eurypterid. Moreover, it is apparent that the occurrences are in all cases immediately associated with the physical and faunal evidences of a break in the series between beds of Lower and Upper Ludlow age, and that this is essentially the horizon at which the eurypterids and Palæophonus nuncius are found on the island of Gotland, marking in both cases what seem to be widespread river deposits which precede the renewed encroachment of the sea in Upper Ludlow time.[12]
9. UPPER SILURIC OF PODOLIA AND GALICIA
Along the Dniester and its tributaries in Galicia and Podolia Upper Siluric rocks have been found containing a few fragments of Eurypterus fischeri. This discovery was one of the earliest and was made by Major-Ingenieur Bloede who found a single impression in a piece of shale from an unknown locality in Podolia. Graf Fischer de Waldheim described this form as Eurypterus tetragonophthalmus, communicating his description to the Société Impériale des Naturalistes de Moscow in 1839 (64). The specific name was given because the eyes were supposed to be of a tetragonal outline, but subsequent study showed that they had the typical margins, and the form was later identified first as E. remipes, then as E. fischeri. Schmidt records finding the eurypterid remains at the base of the Upper Siluric and notes that just as in the occurrences on Oesel so in Podolia the eurypterids and fish remains are found without any other associates. In regard to the occurrence of E. fischeri noted by Barbôt, Malewski, Alth and others, Schmidt makes the following remarks: "In Podolia occurs a species absolutely identical with ours which was formerly identified with E. remipes, and which will probably make possible even further differentiation from the American species. So far as I know there have been but three undoubted specimens found up to this time: (1) the original specimen of Fischer (now in Moscow) from Zwilewcy on the Smotricz, (2) Bloede's specimen (in our Bergakademie Museum) from Balagowa on the Dniester, (according to Barbôt); and (3) that from the Kiew Museum obtained from Dumanow. Malewski also cited Zawalje, Kitaigorod and Studzienica; but I cannot hold these statements as very reliable, since the specimen from Studzienica which is before me, is the horizontal section of a large Cornulites serpularius (Sil. Syst.) which species is well known to me from Oesel (Johannis)" (Schmidt, 245, 13, 14). The Pterygotus fragments which have been reported, Schmidt considers as identical with P. osiliensis (formerly called P. anglicus) from Rootziküll. Schmidt continues: "Of the latter I know practically every single piece, but I have never found a complete individual. Also in the transition beds from limestone to sandstone at Zalesczyki I have found broken pieces of shell, which, however, deserve no particular further examination (245, 13)." In another place, referring to this last mentioned occurrence he makes the following significant statement: "The uppermost beds at Zalesczyki become sandy and red and the fish alone are present besides the rare Pterygotus" (245, 9). (See sketch map, fig. 14.)
While Alth's paper is undoubtedly excellent for the general stratigraphy and palaeontology of Galicia and Podolia, involving as it does not only the results of his own studies but also those of the earlier investigators, it yet fails to give just the details which are essential for the problem in hand. It helps us very little to know that the eurypterids and a large number of the fossils are found in beds some
 Fig. 14. Sketch Map of Parts of Galicia and Podolia, Showing Localities where Eurypterids have been Found |
25–30 feet thick; the important fact is whether or not they occur in thin bands, isolated from the remaining fossils as is the usual way. It must in fairness be stated that Alth's section on the Upper Siluric beds, or as he called them, the "compact and bituminous limestones," does not pretend to be more than a resume of the important but little known works by Barbôt, de Marny and Malewski, written in Russian, but now made available through this careful German summary. Schmidt's statement has shown that the mode of occurrence above referred to as normal holds also in this case for Pterygotus, but there are many other occurrences for which no data are available. However, the rarity and the poorness of preservation of the fragments which have been found make it a matter of no great importance whether the eurypterids are intimately associated with the marine forms or not. I shall here list the localities as given by Alth and other workers in the same field.
Eurypterus fischeri. 1. Eichwald reports this in a black, compact limestone with corals from Kamieniec podolski, Podolia.
2. Malewski reports it from the same limestone at Dumanów, Zawale and Studzienica, Podolia (see Schmidt's remarks above, p. 148).
3. Fragments reported by Wenjukow from Dumanov, Podolia.
4. Siemiradzki reports this species from Zalucze on the Smotrycz river, in a yellow marly limestone (263, 215).
Pterygotus osiliensis. Exact locality and horizon not given for Alth's specimens.
Schmidt reports this species from transition beds of Zalesczyki, Galicia.
Siemiradzki has also recorded the finding of fragments which seem to be similar to Alth's undetermined Pterygotus sp. in the olive green shales from the same locality (263, 215).
Stylonurus sp. Alth. 1. A tail spine from the plattenkalk of the Borszczower beds from Zamuszyn (Alth Plate V, fig. 4).
2. A tail spine from the light greenish-gray marls into which the gray limestones of the Skalaer group pass upwards, found in the region of the Zbrucz Valley opposite to Zajaczki and north of Husiatyn. (Alth Plate V, fig. 5).
3. A tail spine from the thin limestones opposite Zamuszyn (Alth Plate V, fig. 6).
These occurrences in Galicia and Podolia follow the general rule of being very fragmentary and isolated. From the literature one cannot tell whether the eurypterids were actually found in the same beds with the undoubted marine forms or not.
Pterygotus sp. ind. Siemiradzki. Siemiradzki has reported the occurrence of a telson of an undeterminable species of a Pterygotus from the Devonic coral limestone of Skala, Galicia (263, 215). This is mentioned here to complete the survey of the Austro-Russian occurrences.
THE LUDLOW OF ENGLAND AND THE LUDLOW AND LANARKIAN OF SCOTLAND
Introduction. Although the Siluric of southern Scotland is characterized by large eurypterid faunas at successive horizons, the rocks of the same age in England and Wales, where more open marine conditions obtained, have yielded only two or three fragments, except in the higher Ludlow beds which mark the transition to the continental deposition characterized by the Old Red Sandstone. Even in the Ludlow the remains which in some strata are abundant show a much poorer preservation than do those from Scotland. Complete individuals are never found and although it is possible from the fragments to determine that different genera are represented, more precise identifications are difficult, and in most cases species have been erected simply in order to have some way of designating the fragments belonging to the various genera. In order to understand why all of the specimens from the Siluric of England are so much more poorly preserved than are those from Scotland, it will be necessary briefly to trace the geological changes which were taking place throughout Great Britain during the later Siluric.
In Scotland the Upper Siluric is marked by the approaching continental conditions as evidenced in the deposits of greywackes and flagstones, some barren, some containing a sparse marine fauna and others only fish and eurypterid remains. The conditions as yet were unstable, showing the alternate dominance now of river-borne sediments and now of shore deposits. To the south, however, the sea still covered most of England, though the muds pouring in from the land had made conditions unfavorable for many of the forms of life which thrived in that region during Wenlock time. Thus the corals no longer built up great reefs and only a few survived in the stifling muds wherein the graptolites were buried in such abundance. Brachiopods and the majority of molluscs likewise decreased in number as the migration to more favorable waters to the south progressed. Toward the top of the Upper Ludlow rock in England many rill and ripple-marked sandstones are to be found, some of which show trails. Near the top of this series too, occurs the "Bone-bed" which varies from inch to 6 inches in thickness and is made up largely of fish, eurypterid and crustacean remains, while a few brachiopod shells have been found in places. Geikie has estimated that this bed probably covers an area of over a thousand square miles and yet it never exceeds and seldom reaches 1 foot in thickness.
Following the Upper Ludlow in England comes a series of formations of no very great thickness which has been subdivided into the Tilestones, Downton Castle sandstones, and Ledbury shales. Murchison applied the name "Tilestones" to the whole of the flaggy upper parts of the Ludlow, and since many of the beds are red he included them in the Old Red Sandstone. They were believed to mark a transition period between the Upper Siluric and the Lower Old Red, but to be more like the latter with which they were therefore classed. The Downton sandstones are a group of red, yellow and gray micaceous rocks from 80 to 100 feet thick, occurring in the neighborhood of Downton Castle, Herefordshire, and also supposed to mark a transition period. They are undoubtedly indicative of the regressive movement of the sea, which began in Lower Ludlow time in Scotland but which was not strongly felt in England till the end of the Upper Ludlow. Then in the Downtonian and other "passage beds" were washed into the deposits, terrestrial and lycopodious, vegetal remains, together with eurypterids, Ceratiocaris and vast numbers of Beyrichia kloedeni, together with Lingula cornea and Platyschisma helicites.
In Scotland all of the beds above the Upper Ludlow are called, by the Geological Survey of Great Britain, the "Downtonian." This series is to be looked upon as "stratigraphical equivalents of the Tilestones, Downton sandstones and Ledbury shales which, in Herefordshire, overlie the Upper Ludlow Rocks and have been classified as forming the highest subdivision of the Upper Silurian rocks" (215, 568). It is evident that such a usage of Downtonian will lead to endless confusion, for not a little misunderstanding has already arisen because some authors have placed the English Downton beds in the Lower Old Red, and others have used Downton and Passage Beds almost synonymously. If an attempt is made to use Downtonian in a comprehensive way, coordinate in importance with the terms Wenlock and Ludlow, then difficulties will arise and much circumlocution will be necessary to explain whether the Downton of England or the Downtonian of Scotland is meant and in correlation difficulties will come about because so many different deposits are known by the same name. And especially does it seem inadvisable to adopt a name which in England is used for a subdivision of the Ludlow, and make it in Scotland of the same rank as Ludlow. Therefore, the author most fully agrees with the suggestion made by Goodchild that all of the rocks above the Upper Ludlow in Scotland be hereafter designated by the term Lanarkian from the locality in which those higher Siluric beds are so well exposed.
The Lanarkian is a series of conglomerates and sandstones with a total thickness of about 2800 feet, which are either unfossiliferous or contain only fish and eurypterid remains with the usual ostracods, and with Dictyocaris and Ceratiocaris. Plant remains, a myriopod and a scorpion are among the local associates of the above fauna. The series in Scotland is a more strongly marked continental one than that in England. Thus there was a gradual retreat of the sea from the north towards the south, beginning in Scotland in Lower Ludlow time, if not earlier, and leaving all of England except Devonshire dry by the end of the Siluric.
The Upper Siluric of England. It is only the higher divisions of the Ludlow in England which contain eurypterids: i. e., the Upper Ludlow rock, the Ledbury shales, the Downton Castle sandstone, and the Tilestones, according to the commonly accepted classification. Elles and Slater, who have done a great deal of work in the Ludlow district, have been able to determine smaller subdivisions; and since it is from these horizons that the eurypterids have been obtained, I quote so much of the new classification as is needed to follow the merostome occurrences (61, 198).
| III. | Temeside Group | F. Temeside or Eurypterid shales. | |
| E. Downton Castle or Yellow sandstone. | |||
| II. | Upper Ludlow Grupp | D. Upper Whitcliffe or Chonetes flags. | |
| C. Lower Whitcliffe or Rhynchonella flags. | |||
| I. | Aymestry Grupp | B. Mocktree or Dayia shales. | |
| A. Aymestry or Conchidium limestone. |
A few typical sections summarized from those given by Elles and Slater will serve to bring out the relations between the eurypterid-bearing beds, and the strata containing other groups of organisms. On the right bank of the Teme River near Ludlow Castle a section is exposed showing the beds from the Aymestry Limestone through the Downton Castle sandstone. Just a little south of Dinham Bridge which crosses the Teme, less than half a mile west of Ludlow, the Aymestry limestone is seen. This is characterized by entamerus knighti, Encrinurus punctatus and other typical forms. This massive limestone is succeeded by the Mocktree or Dayia navicula shales, which in turn are followed by the Lower Whitcliffe flags with abundant Rhynchonella nucula, Orthis lunata and more rarely Chonetes striatella. The latter fossil becomes dominant in the Upper Whitcliffe flags which, together with the succeeding beds, are well exposed in the famous Ludford Lane now known as the Whitcliffe Road section, near Dinham Bridge. Chonetes striatella "literally swarms" in these flags, and Orbiculoidea rugata and Orthoceras bullatum are likewise prolific. It is these flags which show the first traces of fragments of that little known eurypterid, Pterygotus problematicus, which occurs in the thin shales and sandstones with Spirifera elevata, Chonetes striatella and Orbiculoidea rugata (see fig. 15). The shales and sandstones carrying the fauna just mentioned, are only four feet thick, yet eight changes in sedimentation are shown, marking a rapid alternation of mud and sand deposition which is clearly indicative of near-shore conditions. Immediately overlying this series is the topmost member of the series, the Ludlow Bone-Bed which though never more than a foot in thickness, is yet one of the most noted of the formations of Britain. It has been the subject of description and speculation for seventy-five years or more; but, so far as I know, its origin has never been satisfactorily accounted for (see proposed explanation below, p. 158). Elles and Wood describe the appearance of the Bone-Bed in this section as follows: "It is best developed at the lower end of the section, on the south side of the road where it is 2 feet above road-level, and reaches a maximum thickness of nearly 6 inches. It is, however, very commonly separated into two thin bands of 'bony' material, divided by a few inches of soft mudstone. These bands occur in a more or less lenticular manner, and one or the other disappears almost entirely from time to time, even within the short distance occupied by the section (72 yards). This feature is characteristic of all the bone-beds of these highest Silurian rocks. In addition to the numerous fish-remains and crustacean remains which the Bone-Bed contains, we have identified Chonetes striatella, Orbiculoidea rugata, and Orthis sp: a similar fauna, with Beyrichia in addition, being found in the softer mudstone separating the 'bony layers.' " (61, 203).
Above the Bone-Bed there is a physical and faunal change, the sediments are coarser, sandstones predominating, with only thin interbedded shales, while the genera of brachiopods so characteristic of the strata of the Aymestry and Lower Ludlow have almost vanished with the exception of the Lingulae. Life on the whole became scarce, only the fish, crustacean, and eurypterid remains occurring in any abundance, and these, as is customary, only at certain horizons. The mottled sandstones and shales immediately overlying the Bone-Bed and forming the base of the Downton Castle sandstones is practically barren. Then follows a thin band with Beyrichia which gives place to the Platyschisma bed proper (E b) which is composed almost entirely of Platyschisma helicites and Modiolopsis complanata. This band is delimited upwards by a second Beyrichia zone. Finally, the massive, yellow, micaceous sandstones of the typical Downton appear (E c). These show leaf-like shale partings with Beyrichia, and other beds with fragments of eurypterids together with the plant (a spore?) Pachytheca, and with Lingula minima.
The Temeside or Eurypterid shales (F) are not seen in the Ludford Lane section in which even the Downton group is incomplete. It is not possible to find a continuous section at any one place; but the contacts between each pair of the groups have been seen, so that by combining the sections exposed within a distance of about four miles the entire sequence may be obtained. The contact between the Downton Castle sandstones and the Temeside shales may be seen at Forge Bridge, a little over half a mile northeast of Downton Castle; and the junction between the Temeside group and the Old Red sandstone is visible at Tin-Mill Race about half a mile beyond Forge Bridge. The contact of the lowest division of this group (Fa) with the underlying Downton Castle sandstones is not here observable. The first beds are rubbly shales which, a short distance up, contain a band of red shale. At a higher level occurs a local bed containing broken Lingula cornea, Onchus teniustriatus, Ctenacanthus-like spines and Leperditia cf. marginata (61, 211). There follows a thin sandstone bed, and then a grey shale with Lingula cornea, above which comes the typical oliveshale of F d with the Temeside Bone-Bed which is very similar to the Ludlow Bone-Bed. From this horizon Elles and Slater record the following interesting fauna (61, 211):
- Pterygotus ludensis
- P. problematicus
- Onchus teniuistriatus
- O. murchisoni
- O. sp.
- Lingula cornea
- Ctenacanthus sp. (?)
- Cephalaspis sp. (?)
- Pachytheca sphaerica
The olive shales above the Bone-Bed also contain many fragments of eurypterids. Evidence of the approach of the Old Red sandstone deposition is seen in the frequent occurrence of grit bands in the olive shales. The top of F f is the "Fragment-Bed" which is crowded with fragments of carbonaceous material whose origin is uncertain, and this layer is everywhere succeeded by the purple-red sandstones of the Old Red.
These sections show the typical lithological and faunal characteristics of the Ludlow in England, and they offer unquestionable evidence for a change from marine to continental conditions of sedimentation. Beginning with the Aymestry group which is a pure marine limestone in the lower part, passing up into shales with thin limestone beds, the succession continues through the flags of the Upper Ludlow group, terminated by the Ludlow Bone-Bed, and finally the Temeside group closes the Siluric. These last beds consist of the Downton Castle sandstones in the lower half, which show an alternation of unfossiliferous sandstones and shales with beds of similar character bearing Lingulas or Platyschisma, or eurypterids or fish remains, while the upper portion constitutes the Temeside or Eurypterus-shales which are dominantly eurypterid-bearing, olive shales, with intercalated grit bands, fish beds and bone-beds. In regard to these formations in the Ludlow-Downton district, Elles and Slater make the following significant statement: "Palæontologically, these rocks are characterized by the presence of Eurypteridae, which, although rare in the lower beds, gradually increase in importance until they attain their maximum development in the beds immediately underlying the Old Red sandstone. The rich brachiopod-fauna, characteristic of the lower beds, dwindles and almost dies out with the approach of shallow-water conditions, although the molluscs are somewhat more persistent" (37, 197). The eurypyterids occur in thin seams not associated with the fast diminishing marine fauna, but with crustacea such as Beyrichia, with the thin-shelled Platyschisma helicites and occasional Lingulas, and especially with fishes. The eurypterids are scarce in the Aymestry and Upper Ludlow groups, but become abundant in certain layers in the Temeside group where they are found in cross-bedded sandstones, in bone-beds, and characteristically, in olive-colored shales.
The physical and faunal characteristics which have just been described have usually been interpreted as indicative of shallow-water marine conditions of sedimentation during the late Siluric. The physical nature of the Temeside group would, perhaps, not preclude such an origin, but the faunal characters leave no doubt that the Temeside group must have been deposited on the land. As is so frequently the case in such successions of sands and flags, there is no doubt that the material is terrigenous in origin, the only question being the place of deposition, whether on the land or along the littoral margin of the sea. If we apply the criteria for the recognition of the various types of fossil faunas, it is at once evident that neither throughout the Temeside group nor at any particular horizon in it is there a marine fauna, for we have seen that a marine fauna, whether existing under the rather uniform conditions of the open sea or under the more variable environment of the littoral zone, and whether in a sandy, muddy, or pure water facies, was yet made up of diverse classes of organisms with a scattering representation through the phyla of the Invertebrata. In the Temeside group there is no bed containing representatives of more than two invertebrate phyla and usually only one phylum is represented. The maximum thickness of this group is 170 feet. In the lower 50 feet (Downton Castle sandstones) the deposits are cross-bedded and contain Lingula minima at certain levels, but no other fossils. Such a series is to be accounted for only by deposition at the mouth of a river, either on the subaërial portion of a delta or on the flood-plain, but the coarseness of the deposits implies that the former was the more probable region of deposition. The presence of beds of Lingulae is easily accounted for by the nature of the shells which are thin, corneous, and consequently of small specific gravity. Exceptionally high tides would easily wash in such light shells far up over the delta, while heavier shells would be dropped farther out in the littoral waters. It is evident that the Lingulae must have been transported from their original habitat since they are unassociated with any other forms of life, unless they can be regarded as living in the river mouths. Thus the assortment seems to have been by specific gravity. In the 120 feet of the Temeside group, Lingula cornea replaces L. minima in the single bands, and is to be accounted for in a similar manner. Now it might be suggested that the eurypterids, which are likewise found in thin bands, were also washed in from the sea on account of their light specific gravity; but the difference between the two cases is that the Lingulae are found in abundance in the marine littoral fauna where they occur, normally associated with marine species of molluscs, crustacea, etc., in the marine deposits of the same age further south; furthermore, Lingulae are found from the Cambric to the present in undoubted marine associations. The eurypterids, on the other hand, are not found in the unequivocal marine deposits to the south, but appear quite as suddenly as the Lingulae, although in separate bands. They have been found to the north of the Ludlow area in Scotland, always as concomitants of the transition from, marine to continental conditions, and it is only when the latter conditions transgress farther and farther south that the eurypterids appear.
I think that much light will be thrown upon the interpretation of the late Siluric deposits in England by the study of the Ludlow and higher bone-beds. It will not be possible in this paper to consider the habitat of the early fishes except incidentally, but if that is proved to be fluviatile, as I think it may be, then the following explanation may be offered for the bone-beds. The Ludlow Bone-Bed, which is the most constant and widespread, appears to mark the wholesale destruction of the fishes in the rivers at the time when, in the oscillatory movements preceding and accompanying the retreat of the sea, there were temporary advances. The salt water, pushing its way up the rivers, killed the fishes and other river organisms in great numbers, for the fluviatile fishes can less easily survive an influx of salt water than marine fishes can an influx of fresh water. This is implied in Günther's statement that "On the whole, instances of marine fishes voluntarily entering brackish or fresh water are very numerous, whilst fresh-water fishes proper but rarely descend into salt water" (97, 187). Thus during the oscillations preceding a negative eustatic movement, the sea would occasionally advance a short distance over the land, and if this temporary positive movement were widespread, bone-beds would be formed at or near the mouths of many rivers almost contemporaneously, and even if some areas were submerged and others not, geologically the bone-beds would appear to be approximately synchronous. This theory is borne out by the occurrence of thin bone-beds at a number of higher levels in the beds above the Ludlow Bone-Bed. Moreover, whenever there was a slight retreat of the sea with the pushing forward of terrigenous, coarse material, then the light Lingula shells might well be left stranded along the line marking the high-water level for that particular period. If such a negative movement were followed by a slight positive one, with the consequent killing off of the fish, a bone-bed would be formed and in a given section would be found overlying a Lingula bed, as does the Temeside Bone-Bed (F d). Were the sea to retreat again, more Lingulae would be left stranded, while fluviatile organisms that were light enough might be floated out across the flood-plains of the rivers. These flood plains had but just been retrieved from the sea and would have been so slightly raised above sea level that only lighter organic remains such as the Lingulae were washed over it, thus fluviatile remains of small specific gravity would be carried out across the flood-plain there to come to rest with the Lingulae, and in this way the olive shales with eurypterid fragments and Lingula cornea would be easily explained. The impossibility of considering either fish or eurypterids as washed in from the sea is indicated by the absence of these forms in the open marine waters to the south. While I have made no attempt to prove the fluviatile habitat of the fishes, yet the bone-beds seem capable of explanation on no other hypothesis. Sometimes the beds are only inch thick, containing no complete remains but only a great mass of broken bones, spines, and scales. Such an accumulation could be formed only of transported material, the fish skeletons having been entirely scattered. If a bone-bed were accounted for as due to the sudden destruction of fishes in the sea by a current of colder or more saline water, by an earthquake or some other catastrophic calamity, then the fish would die in great numbers, but their remains would be buried in situ. An illustration of this is found in the case of the tile fish off the New England coast, where, in 1882, according to estimates, over one billion fish were destroyed, and the ocean floor was covered in this region to a depth of 6 feet with the bodies of the dead tile fish (Grabau, 87, 195). Entire skeletons would be preserved in the rapid burial, and other marine organisms which suffered the same fate as the fish would also be entombed, so that the resulting deposit would in no way resemble the bone-beds, which are made up of fragments, usually so broken that identification cannot be made, while marine shells are only rarely found.
The Ludlow and Lanarkian of Lanarkshire. The inliers of Siluric rocks are larger in Lanarkshire than in the Pentland Hills, and the succession is shown more completely, for in Lanarkshire the structure is anticlinal, while in the Pentland Hills the beds have been repeatedly faulted and stand nearly vertical, making it impossible to trace an outcrop except along the strike. About 5500 feet of Siluric strata are exposed, ranging in age from questionable Wenlock, through the Ludlow and Lanarkian (Downtonian) and into the volcanic series of the Lower Old Red sandstone. The eurypterids are found in many more localities than in the Pentland Hills, but they are never so abundant nor are so many genera and species represented. There are four important Siluric areas in Lanarkshire, but in only two of these have eurypterids been found, namely in (1) the Lesmahagow inlier, and (2) the anticline of the Hagshaw Hills.
(1) The Lesmahagow Inlier. This is the larger of the two anticlines and extends from a little north of Muirkirk northeast for 6 miles. The Greenock Water in the southwest and the Logan Water in the northeast have exposed a number of excellent sections in the gently dipping beds. The lowest beds exposed consist of a series of blue greywackes with shale partings, the whole comprising 1300 feet as seen along the southern margin of the area along the headwaters of the Ponesk and Nethan. Only a few specifically unidentifiable fossils have been obtained from this series which is provisionally placed with the Wenlock. Immediately to the north of these beds occur grey, blue and olive shales, with occasional nodular greywacke bands yielding a good representation of lowest Ludlow fossils.
The third subdivision recognized by Peach and Horne constitutes the so-called Ceratiocaris beds which are of particular significance because of the surprising abundance in some places of several species of Ceratiocaris, and because of the occurrence of the Ludlow fish, Thelodus scoticus in one layer and finally because of the association of eurypterid remains with both of these. At many different points along the Logan Water the beds are excellently shown. In a small gorge about three-quarters of a mile to the northeast of Logan House the lowest of the Ceratiocaris beds dipping to the northwest are succeeded by some zones of dark, fissile calcareous flaggy shales which weather a rusty brown and which have yielded the following fossils:
- Worm tracks
- Ceratiocaris laxa Woodw. and Jones
- Ceratiocaris longa Woodw. and Jones
- Ceratiocaris papilio Salter
- Ceratiocaris stygius Salter
- Ceratiocaris telson, like murchisoni M'Coy
- Slimonia acuminata Salter
In the same shale band about a half a mile distant the following fossils were found, the eurypterids occurring in great abundance, but the Ludlow fish Thelodus scoticus being represented by only two fragments (215, 573):
- Myriopoda ? (impressions of)
- Ceratiocaris sp.
- Dictyocaris ramsayi Salter
- Pterygotus bilobus Salter
- Slimonia acuminata Salter
- Thelodus scoticus Traq.
In certain members of the Ceratiocaris group, though a little below the fish horizon, there are recorded from Long Burn, a tributary of Logan Water, the following species:
- Modiolopsis nilssoni (His.)
- Spirorbis sp.
- Beyrichia kloedeni (M'Coy)
- Beyrichia kloedeni var. torosa (Jones)
- Lingula minima (Sow.)
- Orthonota sp.
- Ceratiocaris
- Dictyocaris ramsayi (Salt.)
- Pterygotus bilobus (Salt.)
- Platyschisma (Trochus) helicites (Sow.)
The best development of the Ludlow fish band occurs about mile south of Logan House in which place also was found an excellently preserved scorpion Palaeophonus caledonicus. In the same place in a cliff about 30 feet high a good section is exposed, showing hard greywacke bands at the top, but below these are brown flaggy shales containing Ceratiocaris in abundance and a few Pterygotus fragments. Embedded in these shales are ironstone nodules which contain fish remains. From this outcrop the following fossils have been collected (215, 574):
- Archidesmus loganensis Peach
- Ceratiocaris longa Jones and Woodw.
- Ceratiocaris murchisoni ? M'Coy
- Ceratiocaris papilio Salter
- Ceratiocaris slygius Salter
- Slimonia acuminata Salter
- Physocaris sp.
- Pterinea retroflexa Wahl.
- Platyschisma (Trochus) helicites Sow.
- Thelodus scoticus Traq.
- Thelodus planus Traq.
- Fish fragment undetermined
Overlying the Ceratiocaris beds and appearing as a narrow band to the north of them throughout the area is a series of hard blue and grey flaggy shales and mudstone, with occasional calcareous nodules. These constitute the Pterygotus beds, 350 feet thick, and are the ones from which Dr. Slimon of Lesmahagow made his extensive collections. The best section is along the Logan Water which for quite a distance runs along the strike of the beds. On the right bank about 400 yards west of Dunside the following fossils have been collected (215, 575):
- Ceratiocaris papilio (Salt.)
- Neolimulus falcata (Woodw.)
- Eurypterus lanceolatus (Salt.)
- Eurypterus obesus (Woodw.)
- Eurypterus scorpioides (Woodw.)
- Pterygotus bilobus (Salt.)
- Pterygotus bilobus var. acidens (Woodw.)
- Pterygotus bilobus var. inornatus (Woodw.)
- Pterygotus raniceps (Woodw.)
- Slimonia acuminata (Salt.)
- Stylonurus logani (Woodw.)
- Lingula minima (Sow.)
"In a small tributary of the Logan Water from the north, at a spot about 250 yards west from Dunside, these flaggy shales have yielded specimens of Spirorbis lewisi, Beyrichia kloedeni, Dictyocaris slimoni, Pterygotus bilobus, Slimonia acuminata and Platyschisma helicites."
In several others of the tributary burns the eurypterids are found associated always with Ceratiocaris or Dictyocaris, Beyrichia kloedeni usually and sometimes lingulas.
Still higher horizons of the Ludlow, numbers 5 and 6 of Peach and Horne's subdivisions have yielded eurypterid remains in the basin of the Greenock Water. From the sand greywackes and greenish shales E. N. E. of Waterhead the following fossils are recorded (215, 576):
- Slimonia acuminata (Salt.)
- Beyrichia kloedeni (M'Coy)
- Dirtyocaris sp.
- Spirorbis lewisi (Sow.)
- Goniophora cymbaeformis (Sow.)
- Modiolopsis complanata (Sow.)
- M. nilssoni (His.)
- Orthora impressa (Sow.)
- O. rotundata (Sow.)
- O. solenoides (Sow.)
- Platyschisma helicites (Sow.)
Following upon the highest of the Ludlow green flaggy and sandy greywackes there is in many localities a conglomerate of varying thickness conformable, so it is stated, upon the Ludlow. In the Lesmahagow inlier, however, this conglomerate is absent. On the northwest slope of the anticline the transition beds are exposed in many places showing the change from greywackes to cross-bedded red and yellow sandstones, 1300 feet thick, and constituting subdivision 8. Overlying this is a group of strata, about 100 feet in thickness, containing the very important fish-band. Sections along the Dippal Burn and various streamlets emptying into the Glengarel and Kype Waters show the succession. The fish-band itself is only from 12 to 15 feet thick, comprising an alternating series of brown flaggy carbonaceous shales and green muds tones. It is in the former that the fishes and eurypterids occur, but no organic remains have been found in the mudstones. There are many sections from which the fish and eurypterids have been obtained, but two will suffice to show the nature of the fauna. Near the head of Dippal Burn there have been obtained (215, 578):
- Eurypterus dolichoschelus (Laurie)
- Ceratiocaris sp.
- Lanarkia spinulosa (Traq.)
- L. horrida (Traq.)
- L. spinosa (Traq.)
- Thelodus scoticus (Traq.)
- Birkenia elegans (Traq.)
- Pachytheca
- Parka n. sp.
- Fucoid-like markings
One of the two best localities for ichthyolites and the one in which all of the species of Downtonian fish determined by Dr. Traquair have been found is in the Slot Burn, one of the tributaries of the Greenock Water. The fossils thus far described from there are (215, 578):
- Eurypterus dolichoschelus (Laurie)
- Stylonurus ornatus (Laurie)
- Myriopod
- Lanarkia spinulosa (Traq.)
- L. horrida (Traq.)
- L. spinosa (Traq.)
- Thelodus scoticus (Traq.)
- Birkenia elegans (Traq.)
- Lasanius problematicus (Traq.)
- Ateleaspis tessellata (Traq.)
- Ceratiocaris laxa (Jones & Woodw.)
- Dictyocaris sp.
- Pachytheca sp.
- Plant stems.
- Sponge?
A second fish band yielding several species of fishes and a myriopod has been found a short distance up the Slot Burn and at a slightly higher horizon than the main one; eurypterids have not yet been found in it.
An excellent section in the eastern area of the Lesmahagow anticline is shown in the Birkenhead Burn, a tributary of the Logan Water. The passage from the Ludlow to the Downtonian is obscured by a normal fault which abruptly truncates the Ludlow series, but the rest of the succession is complete. The total thickness of the fish-band with the intercalated mudstones is here fifteen feet. The lowest fossiferous carbonaceous seam is about a foot thick, while higher up in the band the seams vary from one to six inches. "The remarkable feature of this exposure is the constant association of the fish fauna with eurypterids that are characteristic of the underlying Upper Ludlow rocks." The fossils listed are (215, 580):
- Eurypterus sp.
- Pterygotus bilobus ? (Salt.)
- Lanarkia horrida (Traq.)
- L. spinosa (Traq.)
- L. spinulosa (Traq.)
- Slimonia acuminata (Salt.)
- Stylonurus sp.
- Thelodus scoticus Traq.
- Ateleaspis tessellata (Traq.)
- Ceratiocaris sp.
- Birkenia elegans (Traq.)
- Lasanius problematicus (Traq.)
- Plants
- Sponge
2. The Anticline of the Hagshaw Hills. About five miles to the south of the Lesmahagow anticline rises the crest of the Hagshaw Hills anticline, the axis trending northeast southwest. The area between the two anticlines is occupied by a northern belt of limestone, Mississippic in age (Calciferous limestone of Scottish geologists), and by a southern area of Lower Old Red sandstone with one patch of Upper Old Red. Rising above these is the anticline forming the Hagshaw Hills, where the Wenlock, Ludlow and Downtonian are exposed by erosion. It is only in the northern limb of the anticline that the Wenlock and Ludlow are visible, for the southern has been cut off by a thrust fault along the plane of which the older Siluric rocks have been brought to rest against the younger ones. The Douglas Water and its many small tributaries have exposed a number of good sections in the western area of the anticline. One of the best of these is in the Ree Burn, south of the Glenbuck Reservoir where there is an almost continuous section of the Ludlow rocks. At one point in certain blue finely bedded shales and flaggy greywackes specimens of Ceratiocaris, Slimonia and Beyrichia kloedeni, have been found. Along the southeastern slope of the anticline where the Podowrin Burn joins the Douglas Water near Monksfoot a transverse section of the Ludlow rocks is shown. They are greywackes and flaggy shales and are thought to be the equivalent of the lowest Ludlow in the Lesmahagow area. It has not been possible to obtain any definite statement as to the exact horizons in which the fossils occur and whether the eurypterids occur as they do elsewhere in bands distinct from the layers containing molluscs, brachiopods, etc. From this locality the Ludlow beds have yielded the following fossils (215, 583):
- Slimonia acuminata (Salt.)
- Beyrichia kloedeni (M'Coy)
- Ceratiocaris papilio (Salt.)
- Favosites asper (D'Orb.)
- Lindströmia sp.
- Glyptocrinus basalis (M'Coy)
- Crinoid stems
- Ceriopora sp.
- Strophomena (Leptaena) rhomboidalis (Wilck.)
- Ctenodonta sp.
- Cornulites sp.
- Calymene blumenbachii (Brong.)
- Encrinurus sp.
- Illaenus sp.
- Proetus stokesi (Murch.)
- Athyris (Glassia) compressa (Sow.)
- Orthis bouchardi (Dav.)
- O. (Dalmanella) elegantula (Dalm.)
- O. polygramma (Sow.)
- Orthonota sp.
- Orthoceras angulatum (Wahl.)
- O. small smooth sp.
Along the Smithy Burn, the West branch of the Podowrin Burn, Mr. Tait found a brown sandy shale which because of the abundance of the Bryozoan Glauconome has been called the Glauconome band. This immediately overlies the fish beds and contains (215, 585):
- Eurypterus dolichoschelus (Laurie)
- Glauconome disticha (Goldf.)
- Lasanius problematicus (Traq.)
- Ateleaspis tessellata (Traq.)
- Spirorbis sp.
- Sponge
- Pachytheca sp.
The Lanarkian series is typically developed along the northern limb of the anticline from the local conglomerate at the base found only in the Hagshaw Hills to the chocolate-colored sandstones at the top, but the only bed of particular interest in the present discussion is the fish band which has been found in several places. In the Monk's Water, about three-quarters of a mile south of Monkshead, the following fossils are reported from this band:
- Eurypterus, small sp.
- Scorpion
- Ceratiocaris ?
- Thelodus scoticus (Traq.)
- Birkenia elegans (Traq.)
- Lanarkia spinosa (Traq.)
- L. spinosa (Traq.)
- L. horrida (Traq.)
- Lasanius problematicus (Traq.)
- Sponge?
11. THE OLD RED SANDSTONE
History and Subdivision. The closing stages of the Siluric in northwestern Europe were marked by an expansion of the continental areas and an accompanying widespread retreat of the sea which left all of Great Britain except the southwestern portion of Devonshire, all of Scandinavia, Finland, and the northern borders of Germany dry land. Over the region thus exposed was deposited a great series of conglomerates, sandstones, and shales, dominantly red in color, and reaching a thickness of many thousands of feet, the formations being collectively called the Old Red sandstone facies of the Devonic. It was early recognized that the conditions of sedimentation under which these deposits accumulated were essentially different from those under which the marine Devonic limestones of Russia, western Europe, and extreme southwest England were formed. Not only did the tremendous thickness of the beds attract attention, but the coarseness and prevailing red color of the deposits, and particularly the almost entire absence of organic remains, caused considerable speculation on the part of continental as well as British geologists on the origin of this remarkable series. In the early part of the last century the suggestion was made by Dr. John Fleming that the Old Red might have been deposited in lakes. This theory was eagerly taken up first by Godwin-Austen (6) in 1855 and by a host of later writers, each one of whom contributed some bit of evidence, be it palæontological, geographical or stratigraphical, to show that these Devonic red beds were laid down in lakes. The attempt to prove this theory has, as is so often the case in the development of science, led to careful observations by many men, to the formulation of alternative theories and to the collection of a great mass of valuable data. That the first theory may perhaps prove incorrect is of small importance compared to the fact that it made geologists realize that there was a problem to be solved, and spurred them on to its solution. This lacustrine theory, however, has had a longer life than is usually allotted to first theories, for it has held on to the present day and still has more adherents than has any later hypothesis. The monograph by Sir Archibald Geikie "On the Old Red Sandstone of Western Europe," published in 1878, embodied such an elaborate discussion of the various lakes of the Devonic period and so many field observations were adduced to back up the theoretical statements that later writers have with few exceptions considered that the lacustrine origin for the Old Red sandstone was proved beyond any further question. To be sure, one or two heretical geologists have raised objections to these ancient lakes and have preferred to think that the Old Red was a marine deposit formed under particular and inimical conditions. Within the last ten years both of these interpretations have been questioned by not a few, and although the majority of geologists unhesitatingly accept the older ideas, particularly favoring the lacustrine theory, nevertheless, there is an everincreasing tendency at the present time to recognize the fact that all continental formations need not necessarily be deposited in large bodies of standing water. Thus the ultra-modern school of geologists champions the importance of fluviatile deposits in the past, insisting especially upon the fact that such deposits are spread out in large part on the land and not in lakes or inland seas. This school of "terrestrial" as opposed to "aqueous" geologists, found its earliest leaders in Johannes Walther and Albrecht Penck, later disciples in this country being Professors Grabau and Barrell. The last two as well as Walther and Goodchild have argued the dominantly continental origin of the Old Red sandstone, Professor Grabau arguing On the basis of the field evidence and on lithological and palaeontological grounds; and Professor Barrell from the standpoint of the physical conditions which must have prevailed at that time. These various theories will presently be taken up and the evidence for each will be discussed.
The present outcrops of the Old Red sandstone in the British Isles are for the most part discontinuous and decidedly patchy. They fall roughly into five areas (see index map, fig. 16): (1) The Caithness-Orkney
 Fig. 16. Sketch Map of British Isles Showing Distribution of the Old Red Sandstone |
Islands region with a northward continuation into the Shetlands and a southward one into Sutherland and Rosshire, including the coastal strips on both sides of the Moray Firth; (2) The Forfarshire-Kincardine and Perthshire area, with discontinuous outcrops in northern Argylshire, together with patches along the Caledonian Canal; (3) Scattered outcrops in southeast Scotland and the Cheviot Hills; (4) The southwestern and southern district of Wales; (5) western England together with the southern and southwestern portions of Ireland. The lack of geological and geographical continuity in these sections, the distinctness of the faunas where present, and the complicated tectonic relations, have led to many different classifications which have been made to fit not only the facts observed in the field, but also the hypotheses evolved to account for the facts. Moreover, since the deposits have not been formed in the sea, as I shall demonstrate below, none of the usual criteria for correlation of marine strata are available, and thus in each locality where the formations are described local names are given to the beds and it is impossible to state what are the equivalents elsewhere. The same lithological facies are repeated again and again, there being rapid vertical and lateral changes, but nowhere is the succession twice alike.
The original subdivision of the Old Red sandstone was made by Murchison before the middle of the last century into three groups, as follows:
- Upper Old Red or Dura Den beds,
- Middle Old Red or Caithness flags,
- Lower Old Red or Arbroath flags.
The lower series is typically developed in Forfarshire, where it consists of coarse conglomerates for the most part, though shales and sandstone are also represented. The middle series is the remarkable grey, flaggy facies exhibited in Caithness and carrying the abundant fish fauna, while the upper is a yellow sandstone group found overlying the flags at Dura Den. Murchison's chief reason for making the Lower and Middle separate, even though the two are never found in contact or even in the same locality, was the distinctness of the faunas in the two, for while the fish and eurypterids of the Arbroath flags were generically and sometimes specifically like those in the Upper Siluric, they were entirely different from those in the Caithness flags, a statement which later investigators have strengthened. Geikie, however, contended that the Lower and Middle were synchonous deposits in separate lakes and that the faunas were not entirely distinct, and even today Geikie supports the two-fold division making the Lower include the Arbroath flags and the Caithness flags, while in the Upper are the Dura Den beds, which for the most part rest unconformably upon the Lower Old Red or transgressively on older rocks (Geikie, 74, 1006).
Dr. Goodchild, who has worked over the Scottish rocks for nearly fifty years, has taken exception to a number of the prevailing ideas about the Old Red and has given a new subdivision. He has returned to a threefold subdivision for these rocks as they occur in Scotland, the divisions corresponding in many respects to those made by Murchison, although he does not use the terms Lower and Middle because they have been employed with such different meanings by various writers that he deems it best to use locality terms. Thus he gives the following subdivisions of the Old Red sandstone in Scotland, the Orcadian succession being based on Traquair's work (272–275) on the ichthyology and on Flett's studies (66) in the Orkneys (80, 600).
| Upper Old Red Sandstone: | ||
| 2. | Higher subdivision, or Elgin beds (now known to be Triassic). | |
| 1. | Lower subdivisions, or Nairn beds. (Extensive unconformity). |
0–1,000' |
| Orcadian Old Red: | ||
| 5. | John o'Groats Flags. | |
| 4. | Thurso or Rousay Beds. | |
| 3. | Achanarras, Stromness, and Cromarty Beds. | |
| 2. | Berriedale sandstones. | |
| 1. | Badbea Breccias and Basal Conglomerate. | 0–16,000' |
| Caledonian Old Red Sandstone: | ||||
| 3. | Strathmore sandstones (the upper part of which may be contemporaneous with the lowest part of the Orcadian). | |||
| 2. | Myriopod Beds. | |||
| Volcanic Rocks. | ||||
| Acanthodian Beds of Turin Hill. | ||||
| Cephalaspis Beds of Auchtertyre. | ||||
| Volcanic Rocks. | ||||
| Pterygotus Beds of Carmylie, etc. | ||||
| Tealing Beds. | ||||
| 1. | Lower Series of sandstones, mudstones, conglomerates, etc., base not seen. Ranging to ? | 20,000' | ||
(Extensive unconformity).
The Lanarkian Rocks (Downtonians of the Geological Survey, the original Lower Old Red of earlier writers).
Ludlow Rocks.
Mr. George Hickling who has made a special study of the Lower Old Red in Forfarshire, where it is typically developed, has given a somewhat different tabulation (117, 398):
| Feet | |
| Edzell shales | 1,000 |
| Arbroath sandstone | 1,200 |
| Auchmithie conglomerate | 800 |
| Red Head series | 1,500 |
| Cairnconnon series | 2,000 |
| Carmyllie series | 1,000 |
| Dunnottar conglomerate | 5,000 |
| 12,500 |
The employment of different names for deposits perhaps synchronous, but occurring in different localities, is inevitable because of the lack of stratigraphical continuity and because the fossils which are found in these rocks are not of the type to serve as good index fossils, if, as I hope to show, they lived in the rivers.
It will not be possible to work out the lithogenesis of the eurypterid-bearing beds in the Old Red by a study of those beds alone; rather must we take a broader view that will lead to an interpretation of the climatic and other physical conditions which obtained throughout the Devonic in the regions where red sedimentation was going on. Having determined what these conditions were, the origin of the sediments, the agents of transportation and especially the nature of the areas in which deposition occurred, i.e. whether under water or on the land, then the character of the faunas and of the restricted beds in which they occur, will automatically be ascertained. A few detailed sections in the type localities will enable us to generalize later on.
The Caledonian. At the end of the Siluric there was a period of folding and erosion, the extent of which is not known, but most of the sections indicate that it was long, and perhaps nowhere has a true gradational contact been found between the uppermost Siluric and the lower Old Red. Goodchild remarks in this connection, "So far from graduating downward into the Silurian rocks, the local base of the formations under notice (the Caledonian) lies with a violent unconformity upon all of these rocks, and may repose indifferently upon Silurian, Ordovician or even older strata, including the metamorphic rocks of the Southern Highlands of Scotland. What has been taken as the Caledonian Old Red in the cases where it has been supposed that a passage exists is in reality a series of quite different age" (Goodchild, 80, 598, 599). As further evidence of the great break between the two systems Goodchild adds that the Lanarkian rocks shared in all of the tremendous disturbances to which the Siluric rocks were subjected and that "these disturbances had ceased, and had been followed by prolonged denudation, long before the oldest member of the Caledonian Old Red was laid down. Hence it results that the great unconformity, so often referred to, passes above what is left of the Lanarkian rocks. There is no clear evidence of any unconformity below them" (Goodchild, 80, 599).
Thus from the many sections described in the Scottish literature and especially from the authoritative statement of Goodchild, there seems to be good reason for believing that there was a great unconformity at the end of the Siluric, caused in part by profound tectonic disturbances, and that following upon these there was a long period of erosion before the earliest of the Caledonian deposits were laid down. These were of great thickness, amounting in some places to 20,000 feet. As to the origin of the series Goodchild says: "There appears to be evidence of a satisfactory nature that the whole of the vast formation was accumulated under continental conditions, partly in large inland lakes, partly as torrential deposits of various kinds, partly as old desert sands, and partly as the results of extensive volcanic action" (80, 596).
A brief review of the lithological characters and distribution of the Caledonian Old Red series will show most clearly that the rocks throughout are of continental origin. The lowest member, division 1, consisting of sandstones and conglomerates, is often wanting altogether, the overlying volcanics being the first of the series to be present. At the Falls of Clyde, near Lanark, Lanarkshire, these lower beds are, however, to be seen, and they are also found in a few other localities. Generally, the volcanics rest immediately and with a violent unconformity upon various pre-Devonic formations. It is these lavas which are seen in the Ochils and Sidlaw Hills, in the Pentlands and in the vicinity of Oban, at St. Abb's Head and also in the Cheviot Hills. In their greatest development in the Perth and Forfar Hills the volcanics may well reach several thousand feet in thickness, but they thin away toward the north and northeast and pass into alternating sedimentary and igneous rocks which were contemporaneous in their development with the main volcanic outpourings (see sketch map, fig. 17).
The first important fossiliferous beds are those found at Carmylie and adjoining localities in Forfar. These constitute a part of the famous Arbroath flags and because of their abundant eurypterid
 Fig. 17. Sketch Map of Scotland, Showing Localities where Old Red Sandstone Outcrops |
remains are called the Pterygotus beds. In them are found the most perfect specimens of Pterygotus anglicus, though complete individuals are rare, and the rock often contains also an abundance of Parka decipiens, which has been variously identified as crustacean egg cases and as spores of plants. Above these beds follows the main mass of the lavas upon which rest the beds of Auchtertyre, which have yielded Cephalaspis lyelli, Pteraspis mitchelli, and certain of the Acanthodian fishes. At a slightly higher horizon and contemporary with some of the volcanic beds is the Acanthodian zone which is best seen at Tilliewhamland Quarry, Turin Hill, near the town of Forfar. The list of fossils from these beds cited by Goodchild is as follows (80, 507):
- Mesacanthus mitchelli
- Ischnacanthus gracilis
- Climatius scutiger
- C. uncinatus
- C. reticulatus
- Parexus recurvus
- P. falcatus
- Euthacanthus mitchelli
- E. elegans
- E. gracilis
- E. curtus
- Cephalaspis pagei
- C. asper
- Thelodus pagei
- Pterygotus anglicus
- Stylonurus ensiformis
- Parka decipiens
Just above the top of the volcanic series has been found a fossiliferous zone yielding myriopods among which are Kampecaris and Archidesmus, as well as some poorly preserved plants referred to Psilophytum robustum. The top of the Caledonian Old Red is formed by the Strathmore sandstones which are well developed in the Strathmore lowland of Forfar, but the exact age of which is difficult to determine because of the lack of fossils. It has been thought that they might be contemporaneous with the oldest beds of the Orcadian division, but conclusive evidence is lacking.
The Siluric Stonehaven beds of red sandstone and interbedded bright red shales are exposed in the neighborhood of Stonehaven and are about 1500 feet thick. Upon these follows the Dunnottar conglomerate, 5000 feet thick, of coarse red and grey sandstones, grits and conglomerates in which occur pebbles which commonly "range up to a foot or more in length, and yet are astonishingly well rounded. They mostly consist of quartzite" (117, 399). Interbedded lavas occur in the top of the series and are succeeded by the Carmylie beds, about 1000 feet thick, of compact red or grey sandstones with some flags, which are the Acanthodian beds described by Goodchild and which contain the abundant fish and eurypterid remains. This series, together with the contemporaneous lavas, forms the backbone of the Sidlaw Hills. It grades up into the Cairnconnan series of 2000 feet of dull red or grey grit with bands of conglomerate. The succeeding Red Head series, 1500 feet thick, consists in the lower part of "fine red thin-bedded sandstone with bands of hard bright red shale, while the upper portion is made up of thicker-bedded sandstone." Six or seven miles south of the Red Head promontory from which the beds are named, there is a lithological change to blue or grey shales with sandstone partings, illustrating well the rapid lateral variation. Overlying this group is the Auchmithie conglomerate. "The series consists of three main masses of conglomerate, with intervening sandstones and conglomerates. The pebbles in the conglomerates are well rounded, fairly large (generally 1 to 6 inches, rarely 12 inches), and, as usual, are mostly quartzite" (117, 400). This conglomerate is 800 feet thick and is followed by the highest member of the series, the Arbroath sandstone (1200 feet). "Coarse, gritty sometimes pebbly sandstone is its component rock, always red in color" (117 400). The succession as here shown in Forfarshire shows beyond a doubt that the sediments could not have been marine. The complete series is shown in outcrops in Forfarshire, extending over about 500 square miles, while within a distance of less than ten miles the outcrops of all of the formations may be seen.
The Orcadian. Over the greater part of northeast Scotland and extending northward to the Orkney and Shetland Islands there is developed a great series of flags, sandstones and conglomerates younger in age than the Caledonian and these have been called the Orcadian by Goodchild. They constitute the Lower Old Red as used by Geikie and were thought by him to have been deposited in the large water body which he called Lake Orcadie. Neither the natural base nor top of the series has been seen and even the highest members are always followed unconformably by the Upper Old Red. It is unnecessary to take up the formations in detail because they do not contain eurypterids. There are three fossil horizons containing, with one exception, only fish remains. These horizons are the Achanarras beds, the Thurso flags and John o'Groats flags.
Goodchild in summarizing the conditions which obtained in Orcadian time says: "There is evidence that, during the time when the Orcadian Old Red was in course of being deposited, normal pluvial conditions obtained for a time. The deposition of ferric oxide in the old area of inland drainage ceased, chiefly in consequence of the large quantities of vegetable matter which were swept into the old lakes. This latter, in its turn, decomposed the solutions of sulphate of lime, and liberated the calcareous matter, which in a state of diffusion, or aggregated into nodules, now forms so conspicuous an element in the Orcadian Rocks. Furthermore, the sulphate of lime, in its turn, converted the vegetable matter into the bituminoids, which, in a diffused form, permeated—one might almost say saturated—so much of the Caithness Flagstones. I hold, therefore, that the exceptional durability of the Caithness flagstones, which of course is due to the large percentage of bituminous matter they contain, is due to the fact that conditions of inland drainage, one of the phases of desert conditions, prevailed where these occur during the Devonian Period" (80, 220).
Theories of Deposition. From data of the type just given, three theories have been evolved, each based upon practically the same observations in the field, but each involving very different interpretations. The oldest and most widely accepted explanation for the Old Red sandstone is that it is a series of lake deposits; the second theory, which quite rightly has never received very much attention, is that of marine deposition; the newest hypothesis is that the Old Red is dominantly of fluviatile origin and that the deposits were not laid down in any permanent body of standing water, either marine or fresh, but largely on the dry land as torrential and flood-plain deposits or in evanescent playas. I shall briefly consider the first two theories and the objections thereto, and shall then give the third and some of the evidence favoring it. All geologists are agreed that the sediments are clastic, that they were not deposited in the deep sea, that they are land-derived and river-transported; the only point of difference that has arisen is in regard to the locus of deposition.
Deposition in Lakes. This theory has been most fully expounded by Geikie and has been generally accepted in the form in which he gave it. For the British area he recognized five lakes on the basis of the present outcrops, considering that the heavy conglomerates marked the rocky lake shores of Devonic time, while finer deposits pointed out the central portions of the lakes. The presence of desiccation fissures and other structural features to be mentioned below was taken as indicative of the mud flats along shore, which were from time to time inundated by the waters of the lake. Plant, insect, and crustaceous remains, as well as the abundant fish fauna, were correctly pointed out as showing the near presence of land. The distinctness of the fish and merostome faunas in the Caledonian and Orcadian rocks was cited as proof of the distinctness of the lakes in which these organisms had lived. It is surprising, and therefore, worthy of note, that Geikie came so very near to the recognition of the Old Red fish and eurypterids as river dwellers that one marvels at his not having reached that conclusion. The arguments which he cites to account for the differences of the ichthyic fauna of his Lake Orcadie and Lake Caledonia, which were supposed to have been separated by the Grampians, are illustrations taken from modern river faunas; and, if application were made directly to the Old Red faunas, one would have to say that the fish in the two Devonic lakes were different because they came from rivers whose headwaters were separated by a divide. I shall give Geikie's statement in order to show how near he came to the discovery that the Old Red Fauna came from the rivers, and how he failed to realize this because he was so intent on the theory of lakes.
"In the second place," he says, "there does not seem to be any valid reason why the ichthyic fauna of two adjacent but completely disconnected water-basins should not have differed considerably in Old Red Sandstone times, as they do at the present day. Even in the same river-system it is well known that the fishes of the higher portions of the basin are sometimes far from corresponding with those in the maritime parts of the area. Neighboring drainage-basins, divided by a comparatively unimportant watershed, sometimes show a remarkable contrast in their fishes. This has been well pointed out by Professor E. D. Cope, in a suggestive paper "On the Distribution of Fresh-water Fishes in the Alleghany Region of South-western Virginia."[13] The James and Roanoke rivers descend the eastern slope of the continent and discharge into the Atlantic. In their upper waters they have only four species of fish in common. In the upper waters of the rivers Holston and Kanawha, which flow south-westwards into the Mississippi basin, there are only two species alike. Between those eastern and western pairs of rivers runs the more marked water-parting of the Alleghany chain. Out of fifty-six species of fish obtained from the head waters of the four rivers, five were found by Mr. Cope on both sides of the water-shed. There is likewise considerable disparity in the genera represented in the different rivers. The still more important barrier of the Rocky Mountains separates ichthyological areas yet more sharply marked off from each other. Such isolated basins as Lake Baikal, Lake Titicaca, and the Caspian Sea show by their peculiar assemblages of fishes how much ichthyic types may be modified by prolonged isolation. The differences, therefore, between the fauna of Lake Orcadie and Lake Caledonia during the Old Red Sandstone, as I venture to hold, are not incompatible with the idea that the two lakes were in a general and geological sense contemporaneous, though separated from each other by the barrier of the Grampian Mountains, which formed an effectual boundary between two ichthyic faunas" (71, 364, 365).
Deposition in the Sea. To certain geologists it will appear that I am wasting paper in setting forth a theory which has as its thesis the deposition of the Old Red sandstone in the sea, and that it is a further useless expenditure of ink and of the reader's time for me to voice the objections to such a theory. Indeed, I would agree with anyone who raised such a protest were it not for the deplorable fact that there are still not a few geologists who claim that this much-talked-of red sandstone was deposited in the sea, and further that other sandstones with similar striking lithological and faunal characteristics could have been formed nowhere else but in that region where all sediments have been deposited since the world began, namely, in the littoral zone of the sea.
The chief advocates for the theory of marine deposition are Macnair and Reid who brought out two papers in the Geological Magazine for 1896, one entitled "On the Physical Conditions under which the Old Red Sandstone of Scotland Was Deposited" (159), and the other "Palaeontological Considerations on the Old Red Sandstone of Scotland," (160), in which they sought to prove that physical, stratigraphical and palaeontological evidence all pointed to the marine origin of the Old Red. In a few words their interpretation may be summarized: in pre-Devonic time there was a large land-mass to the northwest of Scotland which supplied the material for much of the marine deposits during Cambric, Ordovicic and Siluric time. At the end of the Siluric the sea began to transgress across Scotland and the land-mass was at the same time depressed until by sinking and by marine erosion the whole area disappeared beneath the sea and the Upper Old Red sandstone was deposited over the whole of Scotland. In the words of the two authors mentioned above, "The great mass of this mountain chain, then, must have lain to the northwest of the present Old Red Sandstone area, and we now proceed to show how after this long period of upheaval the mountain mass once more began to sink below the level of the sea, and that gradually the waters of the Old Red Sandstone sea levelled it down to the very core" (159, 109). They consider that all of the deposits were made along shore, but they are then confronted by the problem of the lack of molluscs and other typical marine forms. This absence they thus account for: "The solution of the problem rather lies in the fact that the presence of peroxide of iron in these rocks is inimical to the preservation of fossils with a calcareous test, and that more especially in the case of sandstones, which even when composed of pure sand are well known to be a bad medium for the preservation of molluscan and other similar organic remains" (159, 116).
Objections to Lake and Marine Theories. Each of the two theories given can explain some facts which the other cannot; but, on the other hand, each has very serious faults due in some cases to incorrect observations, in others to the acceptance of prevalent ideas and in others to unjustifiable deductions. Both theories contain elements of truth, but both are open to many objections. These fall into two groups: (1) Physical, (2) Faunal.
(1) Physical, (a) Red color. Within the last twenty years students of sedimentation have clearly shown that it is impossible for a widespread and thick series of red clastic deposits to be laid down in the sea. The red color, as is well known, is due to the dehydration of sediments which were thoroughly oxidized at the time of deposition. Such oxidation cannot take place under water, but only during exposure to the air. It is not to be supposed that the beds were red when deposited, but that only after dehydration had taken place by the lapse of a long period of time, or through the effect of heat from the interior of the earth, or by pressure was the red color taken on. Of course, certain red beds may receive their final working over under water, but such deposits will be of limited thickness and areal extent. For instance, the Bays sandstone (Upper Ordovicic) of Tennessee, Virginia and adjoining regions is a red calcareous sandstone with a maximum thickness of 1500 feet. Throughout most of the formation organic remains are absent, but in the lower beds marine fossils occur abundantly in a few layers of the red sandstones. These fossils are all mollusca and brachiopods, are numerous and wellpreserved, a fact not compatible with the reasoning of Macnair and Reid. As has been fully explained by Grabau, the main mass of the Bays represents an alluvial fan spread out on the land and having its western and southernmost margins extending into the sea. Thus it was possible for some of the highly oxidized sands to be carried out to sea, where they were deposited and where marine fossils were entombed with them. There is, therefore, nothing inherent in potential red deposits to prevent marine shells from being preserved in them; the difficulty lies in the fact that great thicknesses of potential red beds can not be deposited under conditions where it is possible for marine animals to leave their record, because such deposits must be formed on the land. As for the inimical effects of iron peroxide, it need only be stated that the reddest of deposits contain only a small amount of iron[14] and that it is not the amount but the fineness and perfection of dissemination of the iron that are responsible for the color (Grabau, 87, 621). That sandstones made up of grains of pure silica are bad media for the preservation of molluscs is easily disproven, for one need only recall such highly fossiliferous formations as the Oriskany sandstone and the Schoharie grit of the Devonic of New York, or the Miocenic sands and conglomerates of the Vienna Basin. The reason why so many sandstones are unfossiliferous is generally that they were deposited as terrestrial sediments either fluviatile or eolian.
(b) Marine denudation. A second argument advanced by Macnair and Reid is based upon the assumption that the erosion of the Siluric rocks in the Highlands of Scotland was due to marine denudation and upon faulty observations at certain localities. They argue "The marine denudation of the Silurian rocks of the Highlands of Scotland is not in dispute, but Ramsay and Geikie have assumed a subsequent lake or fresh-water denudation." The conformable deposition, however "of the Old Red Sandstone upon the preceding Upper Silurian deposits in the counties of Edinburgh and Lanark, the Welsh area, and in the St. Lawrence basin, precludes any such idea; for from the base of the Upper Silurian to the top of the Lower Old Red sandstone the sequence of these deposits is unbroken. It therefore follows that the denudation of the rocks of the Highland area being marine, the equivalent deposits occurring in the Upper Silurian and Lower Old Red Sandstone are equally marine" (160, 221). The chief objections to this theory of marine denudation continuing from the beginning of Upper Siluric to the end of Lower Devonic time, fall into three groups: (1) Tectonic. The tectonic relations between the Old Red sandstone and the underlying rocks show that there was profound folding at the end of the Siluric, followed by a long period of erosion before the earliest Old Red sediments were deposited; therefore the two series are not conformable as claimed by Macnair and Reid. (See further p. 173 above). (2) Lithologic. (See below, sections (d), p. 182, and (1), (2), (3) on p. 189). (3) Faunal. (See below, section (b), p. 191).
(c) Salt indicative of marine deposition. The argument that the presence of a salt-bearing stratum in the Old Red at one locality is undoubted evidence of the marine origin of that bed, is of no value unless supported by critical data on the chemical composition of the salt and associated salts if any are present, and on the organic content. Too much is now known concerning the continental origin of many and perhaps the larger number of past and present salt deposits for anyone to claim that the sea was always or even commonly the immediate source of the material. Macnair and Reid would make the presence of the salt band an a priori reason for its marine origin, for they say: "We find in the Moray Firth area a large stratum of yellow saliferous sandstone, interbedded with shales containing remains of Old Red sandstone fishes . . . . and we think that but one conclusion alone can be drawn therefrom—that the formation and its contained fish remains were marine" (160, 221). This type of reasoning is delightfully ingenuous and one that is met with frequently; while the authors do not explicitly state any reason why the salt is marine, the reader yet receives the impression that the presence of fish remains carries a strong presumption, and thus we have the pleasing circle: "The salt is marine because associated with fish, and the fish are marine because found in bands interbedded with salt-bearing sandstones." This whole argument would fall to the ground were anyone to show that the fish were fiuviatile, or that the salt could have some other origin.
(d) Thickness of deposits. The recurrence in the same place of thick boulder and pebble conglomerates interbedded with sandstone and shales, all being dominantly red and showing a complete absence of unequivocal marine fossils such as brachiopods, molluscs, crinoids, and trilobites, and amounting in thickness to many thousands of feet proves conclusively that the beds could not have been deposited by an advancing sea, as contended by Macnair and Reid, nor yet in a lake, as Geikie holds. It is not even necessary to point to the red color or to the absence of marine fossils; the thickness and coarseness of the deposits absolutely precludes the possibility of their having been formed in the sea. Macnair and Reid hold that the sea transgressed from the south to the north, but in that case, while there might well have been a basal conglomerate a few feet thick, this would inevitably have been succeeded vertically by finer deposits, sands at first and then muds or limestones as the water became deeper, and the zone of coarse near-shore deposits would have advanced pari passu with the transgression of the sea. Thus it would have been impossible for coarse material to have been deposited in southern Scotland in the Upper Devonic when the sea shore stood two hundred miles to the northwest. Greater obstacles arise if we attempt to have these deposits formed in lakes or epicontinental seas. In Forfarshire, the
 Fig. 18. Section to Explain the Deposition of the Old Red Sandstone in the North of Scotland |
position of "Lake Caledonia," the estimated thickness given by Hickling is 12,500 feet, including the volcanics, or considerably over 10,000 feet of clastic deposits; in Caithness Geikie estimates the series which he supposed to have been contemporaneously deposited in "Lake Orcadie" at 16,200 feet. These two lakes were separated by the Crystalline Highlands, a strip of land about 90 miles broad, which apparently supplied the sediments for Lake Orcadie. The waves of this great lake, which is estimated to have had at its maximum a surface of about 48,000 square miles, cut back into this old mountain chain which was at the same time being denuded by the rivers which brought their loads into the lake. In its maximum developed Lake Orcadie extended from Nairn to the Shetland Islands, the Orkneys representing a sublacustrine rise. The cross section made by Geikie is here reproduced in order to show his interpretation (fig. 18). It is at once apparent that there was not enough dry land to supply the thousands of feet of flagstones making up the Caithness series. It is even more difficult to surmise whence came the material which filled up the mid-Scottish basin or "Lake Caledonia," for it was hemmed in on the west by a narrow ring of hills separating it from "Lake Lorne" in North Argylshire, and on the south by hills along an east-west line through the Firth of Forth, and on the north by the Highlands which were the source of the 16,000 feet of sediments deposited in Lake Orcadie, while to the east the sea covered France. The only other source would be a mountain chain in the present English Channel, but the objections to this are obvious. A natural question that arises often in reading Geikie's monograph, and one which Macnair and Reid most pertinently ask is how outliers of conglomerates on the tops of high mountains, in the very regions which were supposed to have been lake barriers, are to be accounted for. Geikie has proposed that perhaps they represent old fiord-like indentations in the shoreline. This explanation will not serve, however, when such outliers are found on what must have been the very centre of the ridge between Lakes Orcadie and Caledonia, such, for instance, as Macnair and Reid mention at Mealfourvonie just north of Loch Ness in Inverness where an outlier is found 2284 feet above sea-level, and at Tomintoul in Banff, and Rhynie in Aberdeen. The outliers in all parts of Scotland indicate that the deposit was essentially continuous, though varying in lithological character and origin from place to place.
(e) Structural features. The cross-bedding, ripple marks, and other structural features that are cited by some authors as indicative of marine littoral conditions of sedimentation, by others as lacustrine littoral, will be considered below under the third theory of the origin of the Old Red sandstone (p. 189).
(2) Faunal. Attention should be called to certain erroneous lines of argument that have been used and which fall down because based on false premises. For instance, it is impossible to prove that the Old Red sandstone eurypterids were marine by saying that the Siluric ones were and that therefore the Devonic ones of the same genera must also be. First it must be proved that the Siluric eurypterids were marine. To quote once more from Macnair and Reid: "We have . . . . seen no reason assigned why Eurypterids and Placoderms of the same genera, which are marine in the late Upper Silurian, and fishes of the same genera and species which are equally marine in the Devonian of Russia and Central Europe, as well as in the Devonian of North America, should be termed equivocally marine in the Old Red sandstone" (160, 219). It may be remarked that the Devonic fishes of North America here referred to have been shown, from their occurrence and distribution, to be mostly if not entirely fluviatile (Grabau 87, 88).
Macnair and Reid have with great justification brought forward many objections to the "Lake theory" advocated by Geikie, but their logic fails them when they contend that because the Old Red fish and eurypterids could not have been lacustrine forms, therefore they must have been marine.
The river origin seems never to have occurred to these two writers, or else if it did they considered that the same objections were open to it as to the lake origin. One of the arguments which they advance against the lake theory is the difficulty of the origin and distribution of the fish and eurypterids. They argue thus: these forms were present in the Siluric and so it is not strange that they should occur also in the Devonic; "but of the genera Osteolepis, Dipteris, Glyptolepis, and other fishes of the Old Red Sandstone no undoubted plates or scales occur in the preceding formation. The question therefore arises, whence came these highly organized fishes of the Old Red Sandstone? More especially, from what fresh-water region did they migrate? Not only so, but as the same genera of fishes occur in the Devonian of North America and the St. Lawrence basin, we have an equal right to know by what fresh-water pathway of distribution they were enabled to migrate some 3000 miles between one point and another" (160, 218, 219). But surely such facts of distribution should not be distressing; many a case could be cited in the recent fresh-water fish fauna of the same genera occurring more than 3000 miles apart, and with perhaps no related genera in the intervening area. One may mention the case of the genus Umbra, a form so peculiar as to be made the type of a family in which are only two species, these being most closely allied, and yet one occurs in the rivers of the Atlantic states of North America and the other in the Danube system, some thousands of miles distant. Even more remarkable is the genus Scaphirhynchus among the sturgeons, which likewise has two species: one in the Mississippi system, the other in Central Asia. In the same family is the genus Polyodon, with two species only, one in the Mississippi, the other in the Yangtse-kiang. But one need not confine the illustrations to genera which are identical in distant regions; species offer even more surprising examples. Perca fluviatilis, Gastrosteus pungitius, Lota vulgaris, Salmo salar, and many others might be mentioned, inhabiting both the rivers of eastern North America and of Europe. For an extended discussion on migration the reader is referred to chapter V on that subject below, especially pp. 203–7. These illustrations will suffice to show that fresh-water forms can often migrate for several thousand miles, and that through river distribution even the same species may occur in regions widely separated. It may here be remarked that distance is of less significance than time available for migration (see below, pp. 208 et seq.).
Summary. The objections to the marine and lacustrine theories of deposition for the Old Red may be reduced to the single criticism that they are out of date. The theories were helpful attempts toward the solution of one of the big problems in stratigraphy, but in their formulation and working out, their authors naturally followed the ideas which were accepted as correct twenty years ago; that some of these should have been found to need revision is only an evidence of the progress of science. The study of sedimentation is a branch of geology which is even yet not receiving the attention due it, but, nevertheless, the students of lithogenesis are steadily increasing, and there is more being said and written today about the work of the wind and of rivers in the geological past than there was a dozen years ago.
Theory of Fluviatile Deposition. The conditions up to the beginning of Old Red sandstone time have already been outlined and it was shown that there was a progressive retreat of the sea to the south, leaving all of Scotland and most of England a region of dry land subject to the subaërial forces of denudation, the greatest of which are the winds and the rivers. The rivers cutting down into the newly elevated continent carried great quantities of detritus toward the sea. But these were not the rivers of a pluvial climate. They were rather the torrents which carried off the waters from occasional heavy rains such as occur in semi-arid regions. That the climate must have been relatively dry is indicated by the thickness and great areal extent of the Old Red Sandstone, for, as was explained, these deposits must have been thoroughly oxidized at the time of their deposition in order that they might be potentially red. In post-Devonic time, either by age, heat or pressure, those oxidized deposits became red through dehydration. The climate, then, was semi-arid and the rivers of the nature of torrents which could transport vast quantities of material, but which would in most cases drop that material before reaching the sea. This would be brought about because the streams would soon lose their supply of water, for the rains were only periodic and even the water which was collected into streams would be lost by evaporation or by sinking into the ground. Great alluvial fans were spread out, consisting of coarse conglomerates near the source of supply and of sands farther away. During those periods when the infrequent but heavy rains fell, playa lakes undoubtedly were formed, similar to those known to be characteristic in present semiarid regions which have periodically inundated river flood plains. Evidence is not wanting that just such water bodies did form, for Geikie has called attention to certain characteristics in the Thurso flags which admit of no other interpretation. Along the northern coast of Caithness from Castletown to Thurso, a distance of some seven miles along the beach, these flagstones are exposed in great sheets. They consist of "fissile, calcareous, grey, hard flagstones, green, gray and brown calcareous (and frequently bituminous) shales, with thin bands of calcareous gritty sandstone and argillaceous limestone ('calmy limestone'), seldom more than a few inches in thickness. . . . . Even when split into smooth sheets an inch or less in thickness, these hard, tough layers show on their yellow, weathered edges successive paper-like but mutually adherent laminae. . . . . "
A second feature is "the extraordinary abundance of ripplemarked surfaces and sun-cracks. Though these markings abound also in the lower flagstone group, it is here that they attain their greatest development. Surfaces of flagstone or shale, many square yards in extent, are profusely covered with fine ripple lines as sharply preserved as if only today imprinted on the soft sediment. In many places every successive stratum or leaf of rock is thus marked, so that several distinct rippled surfaces may be counted in the thickness of a few inches of rock. It is likewise observable that the rippling is generally close-set, sometimes not exceeding an inch in breadth from crest to crest of the ridges."
Mud-cracks form a third important structure. Geikie says: "More abundant and admirable illustrations of sun-cracks could hardly be found than occur along the coast. Broad, gently-inclined sheets of rock again and again present themselves to view so covered with reticulations as to look like tessellated pavements. It may be noticed that the cracks not infrequently descend through many of the fine laminæ of deposit for a depth of 5 or 6 inches with occasionally a breadth of 3 or 4 inches. The material filling up the interstices abounds with small, occasionally curved pieces of shale. These may, no doubt, be regarded as portions of the upper muddy layer which cracked off and curled up during desiccation, as may often be observed on dried-up pools at the present time. Some pittings, occasionally seen on the sun-cracked surfaces, may perhaps represent rain-drops" (71, 392, 393).
Such characteristics as those just cited have been used by Geikie as proof of the lake shore origin of the beds and by other writers as indicative of their formation in mud-flats along the sea coast. Were it not that such interpretations are offered by the majority of geologists it would be unnecessary to dwell upon the unequivocal interior continental origin of these features. That mud-cracks should be formed over wide areas indicates beyond a doubt the presence of a large body of very shallow water which completely evaporated, leaving the whole surface exposed to the air. Not only that, but the exposure must have been long for the cracks to be 5 or 6 inches deep and occasionally 3 or 4 inches wide. Professors Grabau and Barrell have discussed this subject of ripple marks and sun-cracks over wide areas in such a convincing and logical manner that it need not be taken up in detail here. In his Principles of Stratigraphy Professor Grabau cites the case of the great playa in the Black Rock Desert, Nevada, which forms in a few minutes and covers an area of from 450 to 500 square miles and yet is seldom over a few inches in depth. Russell has described this lake and records that in a few days all of the water may dry up leaving the surface cracked in all directions. "The lake beds then have a striking resemblance to tesselated pavements. . . . . "—the very words used by Geikie in describing the Old Red flagstones! Grabau says: "Taking the areas of mud-crack formation in the order of their magnitude, the playa surface would probably stand first. Here the entire surface for hundreds of square miles becomes mud-cracked, often to considerable depth, on the complete drying up of the temporary playa lake. Here, too, the conditions for the preservation are most favorable. Not only is the exposure a long one, often the greater part of the year, or for many years, and for much of the time to intense heat, but the chances of proper burial are much greater. Wandering sand dunes may thus preserve the record, dust deposits may fill the fissures, or, at the next flood, sands or muds may be swept into them. In fact, the playa or takyr seems to be the ideal surface for mud-crack record, and one is tempted to refer most mud-cracked strata to such an origin. Certainly where fossil mud-cracks penetrate a formation to the depth of 10 feet, as is the case in the Upper Shinarump (Triassic) shales of Utah, it is difficult to believe that they could be formed under other conditions than those permitting prolonged exposure such as is found only in the playas of the desert, where ten years or more may elapse between rainfalls. . . . . If the playa lake exists for some time it may become stocked with certain forms of organisms, especially types whose eggs or larvæ can be transported by wind or by birds. The small crustaceans Estheria, Daphnia, and Cypris are characteristic of desert lakes, the first being found in ponds which are dry for eleven successive months" (Grabau, 87, 707, 603). The nature of the organisms characteristic of such playa lakes is exceedingly interesting in view of the fact that Geikie adds to his description of the lithological characters of the beds in question th e following statement: "Fragments of fish and coprolite are scattered abundantly through most of the flagstones. Some of the calcareous shales are full of Estheria, while traces of plants occur in great numbers, though generally in a somewhat macerated condition" (71, 393). The close correspondence between the description of modern playa deposits and the Caithness flag portion of the Old Red Sandstone series leaves no reasonable doubt that the latter formation was the result of inland drainage in a semi-arid or desert region.
The detailed characteristics of a single series of beds in the Old Red have been taken as an example illustrating the conditions which prevailed, but attention need not be confined to any single part of the formation, for Goodchild has found evidence in all of the divisions of the Old Red to show that desert conditions prevailed throughout all the Devonic wherever this type of deposition obtained. In order not to burden the discussion with a too lengthy description of all of the features indicating desert or at least continental origin for these deposits I shall give a list setting forth the facts already cited and certain additional ones.
Summary of Evidence for Fluviatile Deposits. (a) Lithogenesis. (1) The presence of finely stratified, rippled and sun-cracked flags over an area of many square miles, and at successive horizons, the sun-cracks penetrating to a depth of five or six inches and being at times three or four inches wide, indicates playas or at least broad river flood plain conditions. These features have been noted by Geikie in the Thurso flags (71, 392, 393).
(2) The presence of clay galls in the deep interstices between the sun-cracked prismatic layers in the Thurso flags indicates exposure of clayey surfaces to the air long enough for flakes to be curled up and blown into the cracks. Such a feature might characterize any sun-cracked area, but the depth of the cracks as cited in (1) indicates a play a or a river flood plain.
(3) The basal conglomerate of the Orcadian series has characteristics pointing to the fact that it is made up of material derived from the disintegrated but not decomposed underlying rocks, thus indicating dry climatic conditions during its formation. The conglomerate is too thick to represent the basal conglomerate formed by an advancing sea, even if other characteristics did not preclude the marine origin. In detail the characteristics are as follows: (a) "The blocks vary in size up to as much as a yard, or even more, in length, and consist of gneiss, pink granite, quartz-porphyry, quartz-rock, mica-schist, and other crystalline rocks, with abundance of pink cleavable orthoclase derived from the underlying gneiss" (71, 375). In the Caledonian series the blocks are even larger, Hickling having recorded them up to 8 feet in diameter. (b) In every case the underlying rock from which the conglomerate boulders were derived can be found not far away. "Near the granite they (the boulders) are made up in great measure of granitic debris. Round the quartz rock they are largely composed of that material. The existence of the well-veined orthoclase gneiss is indicated some distance before the underlying rock is actually seen by the abundant fragments of beautifully cleavable pink felspar in the conglomerates" (71, 370). (c) In both of the quotations just given reference is made to the abundant presence of fresh pink orthoclase. Goodchild has likewise referred to the arkoses with unweathered feldspar fragments (80, 219), and has pointed out that they indicate disintegration under semi-arid or desert conditions. (d) The basal conglomerate is too thick to be of any other than fluviatile, more especially torrential origin. For instance at Sarclet, about five miles south of Wick, Caithness, a great mass, 250 to 300 feet high, rises from the sea, the base not being visible. Here "the matrix, red in colour, and less strongly felspathic than towards the south, contains large and usually rather well waterworn fragments of quartz-rock, granite, felspar, porphyry, and red sandstone" (71, 376). On no sea or lake beach is a large boulder conglomerate 250 feet thick ever formed by the action of waves. Along an open coast exposed to the full force of the waves great boulders may indeed pile up, but they will be in a very narrow strip at the foot of the cliffs and will rapidly decrease in size until within but a few feet from shore no large ones will be found and those which do occur will be in only a thin layer wedging out seaward. Moreover, a boulder conglomerate formed along a seacoast would almost certainly be fossiliferous, as I shall point out below. Such a conglomerate might, however, easily be piled up by the waters of the swift and powerful torrents which periodically occur in desert regions. In large basins of inland drainage the rivers flowing down the enclosing mountains bring in great quantities of debris which is coarse and bouldery near the mountains and finer further out. Davis records that "A great part of Persia consists of large basins enclosed by mountains and without outlet to the sea. Long waste slopes stretch forward five or ten miles with a descent of 1000 to 2000 feet, stony near the mountain flanks and gradually becoming finer textured and more nearly level. The central depressions are absolute deserts of drifting sands with occasional saline lakes or marshes" (87, quoted from Davis, 50, 588).
(b) Faunal. Throughout the Old Red sandstone of Great Britain and the continent, typical marine organisms are absent except where this facies interfingers with the Devonic marine facies. The types of life represented in this whole series are few and yet of exceeding interest, since they are among the earliest of land forms, such as scorpions, insects, freshwater crustacea, fish and eurypterids, while the flora, though much poorer than that from the Gaspé sandstone of New Brunswick, yet shows the presence of ferns, coniferous trees and vascular cryptogams. The Caledonian Old Red, which is largely conglomeratic, has yielded comparatively few fossil remains, but in the Pterygotus-or Carmylie- sandstones of Forfar, Pterygotus anglicus has been found associated with Parka decipiens and at a higher horizon Cephalaspis and Pteraspis occur, and still higher the Acanthodian beds of Turin with a good fish fauna as well as Pterygotus anglicus and Stylonurus ensiformis. Thus, in the Caledonian Old Red, a series 12,500 feet or more in thickness, the fish and eurypterids are the only abundant organisms. This single faunal fact would be sufficient, even though all other types of evidence were wanting, to make me say that those two groups of organisms lived in the rivers (see criteria, p. 77 above). In the Orcadian the fauna is more varied. Traquair, who has made such a careful study of the ichthology of the Old Red Sandstone of Great Britain, has established the following fish zones in the Caithness area (272):
| John O'Groats | 
|
Tristichopterus alatus Egert. |
|
Microbrachius dicki Traq. | |
| Thurso | 
|
Coccosteus minor H. Miller |
|
Thursius pholidotus Traq. | |
|
Osteolepis microlepidotus Pander | |
| Achanarras |
|
Pterichthys, 3 species |
|
|
Cheirolepis trailli, Ag. | |
|
|
Osteolepis macrolepidotus Ag. |
This fish fauna is very different from that to the south of the Grampians in Forfarshire, there being no species in common between the two areas and only two genera, Mesacanthus and Cephalaspis, the latter being represented in Caithness by only a single specimen.[15] From this division no eurypterids have been reported.
In Caithness and in the Orkneys and Shetland isles has been found a phyllopod crustacean of a genus which at present lives in rivers and freshwater lakes and playas, namely, Estheria. T. Rupert Jones has described the species E. murchisonia, which is abundant in a "dark grey, tough, fine-grained, sandy flagstone, slightly micaceous, somewhat varying in tint and hardness. . . . . Great numbers of the valves are spread over large surfaces of the flagstone, sometimes scattered sparsely, sometimes congregated in groups, forming films between the layers of fissile stone" (191, 405). Murchison says of this species: "It occurs in certain localities in such numbers as to form layers an inch or two thick, entirely made up of the thin carapaces" (191, 404).
The Old Red sandstone of Lorne has yielded, besides Pterygotus anglicus remains, two species of chilognathous myriopods, Campecaris forfarensis (Page) and Archidesmus sp. described by Peach (214, 83). These are among the earliest myriopods yet known and suggest that the beds in which they were found were formed on land, for if the myriopods had been transported far they would have been destroyed. Moreover, since they had not hard parts to be preserved, they must have been buried quickly. A playa would be the ideal place for their burial, but I do not know enough about the beds in which they were found to state that they were formed in a playa. Macconochie has discovered in these beds plant remains related to Psilophyton, and a fish which Traquair describes as Cephalaspis lornensis (Macconochie 157, Traquair 273).
Geikie calls our attention to what is believed to be "the oldest lacustrine or fluviatile mollusk yet known, Amnigenia (Anodonta, Archanodon) jukesii. This shell has been found in the Upper Old Red Sandstone of Ireland and England, associated with land-plants, (Archaeopteris, Sphenopteris, Bothrodendron, Ulodendron, Stigmaria, Calamites) fishes (Coccosteus) and arthropods (Eurypterus).
12. MISCELLANEOUS OCCURRENCES
We have now completed the discussion of the significance of the eleven most important eurypterid faunas, the ones which it has seemed to the writer offered the most material from which to draw deductions. In addition there is a certain group of occurrences which appear to be able to throw little light upon the determination of the habitat, and they have not been discussed so far, for, if from the best material which we have at hand it can be proved that the eurypterids lived in the rivers from the very beginning of their history, then we need be no more distressed at finding a fragment among marine remains than we are when we find a single leaf or piece of wood associated with brachiopods and molluscs. But, lest the advocates of the early marine habitat of the eurypterids should complain that I pass over lightly the very cases which seem to prove conclusively to them that their view is correct, I shall take up those cases briefly and show wherein they do not prove what they are supposed to; but rather if of any weight at all, indicate that the eurypterids did not always live where their remains were entombed. These remaining instances, then, fall into three groups.
(1) The presence of a single eurypterid fragment or perhaps two or three fragments associated in the same stratum with a typical, well preserved, marine fauna.
(2) The presence of a single eurypterid fragment or complete individual in a stratum barren of other fossils, but immediately preceded and succeeded by strata carrying marine fossils.
(3) The presence of quite a number of fragments in scattered occurrence, but associated intimately with a typical marine fauna.
To the first group belong the following:
- Echinognathus clevelandi, Utica shale, Upper Ordovicic.
- The eurypterid fauna of Condroz, Upper Devonic of Belgium.
- Pterygotus problematicus, occurrence doubtful in Aymestry limestone.
- Eurypterus punctatus fragments, Wenlock limestone, England.
To the second belong:
- Strabops thacheri, Potosi limestone, Upper Cambric or Lower Ordovicic.
- Eurypterus prominens, Clinton.
- E. boylii, Guelph.
- E. microphthalmus, Manlius; Monroe.
- Pterygotus problematicus, May Hill sandstone, Llandovery.
- Eurypterus sp. Wenlock (of Southern Belt, Scotland).
- Eurypterus sp. Wenlock (Girvan area, Scotland).
- Pterygotus australis. Upper Siluric of Australia (Information insufficient, may belong to group 1).
- Pterygotus osiliensis, Pterygotus marl of Gotland.
To the third group belong:
- The Siluric fauna of Bohemia.
- The Lockport fauna of Ontario.
- The Siluric fauna of Podolia and Galicia probably belongs here.
- Pterygotus sp. Siemiradzki, Middle Devonic of Galicia.
The lines of argument for the above occurrences have been stated from time to time, but are scattered throughout the paper. They may be brought together here for reference since so many of the cases are subject to the same arguments. In chapter III the criteria for recognizing the various types of habitats in the past were fully discussed, and will now be of great help in establishing the nature of the habitat indicated by the various eurypterid occurrences given in the three lists above. In the light of the arguments that have gone before, and especially of the discussion on habitats, the following truths may be considered as self-evident or as easily demonstrable.
1. The occurrence of a single fragment, or of two or three fragments, or of a single complete eurypterid in a formation where it is associated either intimately in the same stratum or closely in adjoining strata with a typical marine fauna, as defined on p. 76, cannot be considered as proof that the eurypterid remains are a part of the marine fauna, for the following reasons: (a) it is impossible to explain how any group of marine organisms could have their remains so completely destroyed that but a single fragment should be left; such is never the case with other groups of marine organisms and it is not logical to suppose that the eurypterids should in so many instances have suffered complete annihilation, leaving only one fragment behind to show that they had lived in the sea of that period. It has been suggested that the eurypterids, like modern crabs and horseshoe crabs, were cannibalistic, not only devouring living members of their own family, but also the molted exoskeletons, in this way destroying most of the hard parts which might otherwise have been preserved. This is an ingenious explanation to account for the fragmentary condition of the eurypterids so frequently observed, but when we attempt to explain similarly the appearance in the rocks at a given horizon, of only one fragment, the result is a reductio ad absurdum. For unless we are to believe in a miraculous mutual devouring, such as that which took place between the "Gingham Dog and the Calico Cat" as so vividly described by Eugene Field, we would still expect survivors from the feast. Are we to let imagination run wild and to picture to ourselves a fierce struggle in those ancient seas between the members of the eurypterid family, a struggle which caused the destruction of young and old alike, friends, neighbors, and relatives, until a single maimed, but victorious individual remained? But, if we go so far, we must look at the last scene, must gaze upon the painful sight of that last survivor, demented by his orgies, tearing his own limbs apart and devouring them until—well, we would expect that his jaws and ectognaths would have been the final things to remain, but strangely in the Utica sea it was a claw which remained. It is painful to think of the destruction of the young merostomes in these periodic holocausts, that whole faunas should have perished leaving no descendants, and of the infinite labor Nature must have had to create a new genera and species for succeeding seas! Yet, when the early Palaeozoic periods were past these frightful scenes of wholesale destructon gave way to gentler, more pacific modes of life, so that in the Upper Siluric in central and western New York and on the Island of Oesel we find indications from the fossils that the eurypterids lived amicably to a ripe old age, dying a natural and peaceful death and enjoying a decent and fitting burial in the fine muds of those times. Thus we see again the steady progress in evolution from the early days of barbarism to the later ones of communal altruism. (b) It is impossible to explain the occurrence of one well preserved eurypterid with no other associates, such for instance as E. prominens; for, if the conditions for perfect preservation obtained, then the rest of the eurypterid fauna should have been preserved. (c) If we are to consider that a single fragment of a eurypterid when found in marine strata proves that the eurypterid lived in the sea, then, provided no other proof existed to the contrary, insects, land shells, leaves, logs, spiders, scorpions and other land forms which are often floated or blown out to sea and which are found today thousands of miles from land, and have often been met with in the rocks associated with marine forms would also be considered as inhabitants of the sea. Since the reasoning given on pp. 93-193 has shown that the most significant and important occurrences of the eurypterids point to a fluviatile habitat, then the single special cases should not be cited as proof to the contrary. It is just as if we were to say that, in spite of the many abundant, well preserved floras of the order Fagales known throughout the world in continental beds from the Cretacic to the present, we were forced to conclude that birch and oak trees have always constituted part of the open marine flora, because in some dredging operations today an oak trunk and a number of birch leaves were hauled up one thousand miles from shore. Specific instances of anomalous occurrences have been cited on p. 67, but I shall give one further illustration here to show how little association may mean.
The Upper Devonic sandstones of Condroz Belgium with an aggregate thickness of 22 m., constitute the sandy phase of the Famennian shales of the lower part of the Upper Devonic. They are of interest because of the mixed marine fauna and terrestrial flora found intermingled in them; brachiopods, pelecypods, land forms including ferns, and the fish characteristic of the upper Old Red of Scotland are found associated, and the American genus Dictyospongia also occurs in this sandstone. Since at least part of the fauna is marine, and the flora is terrestrial, the eurypterids might be interpreted either as marine or fresh water forms; but inasmuch as only a few fragments have been found, the more rational interpretation would seem to be that the organisms did not live in the sea. This is further borne out by the fact that as the Upper Devonic beds are traced to the south into Germany they become pure marine limestones, in which no eurypterids have been found, but traced to the northwest they merge into the Old Red sandstone of England and Scotland which contains eurypterids and fresh water fishes. The deposits in Belgium, then, mark the meeting-place of the marine and terrestrial waters as the sea encroached from the south upon the Upper Old Red shore, and for this reason it is impossible from a study of the fauna, flora and sediments of that region alone to arrive at any conclusion as to the habitat of one group of the organisms whose remains are found there. If, for instance, we had no information from other sources regarding the ecology of the pelecypods, it would not be safe to infer that they were marine organisms because associated with brachiopods, nor would it, on the other hand, be fair to assume that they were terrestrial because ferns were embedded in the same strata. The same may be said for the eurypterids; nothing regarding their habitat can be inferred from their appearance in such beds as these sandstones with a commingled marine and terrestrial assemblage of organic remains. In some cases it is possible to take account of more factors, such as the relative perfection of preservation of the various groups of organisms when one, perhaps, shows evidences of transporation and consequent maceration, or again, the relative scarcity or abundance of species and individuals. In the instance of the sandstones of Condroz, I think that it is justifiable to attach importance to the sparse and fragmentary condition of the eurypterids as compared with the abundance and good condition of the other organic remains, and to conclude that probably the merostomes did not live in the region where their fragments finally came to rest.
2. The truth of the thesis of the above paragraphs being accepted, it must be acknowledged a fortiori that a single fragment or even a complete individual in a stratum in which occur no remains of typical marine organisms, intercalated in strata which do, is not the slightest proof that the eurypterid was an inhabitant of the sea.
I may here, as an illustration, give an account of the occurrence of Strabops thacheri, the only known eurypterid from the Upper Cambric or Lower Ordovicic Potosi Limestone of Missouri. In the section near Flat River, St. François Co., Missouri, given by Nason, the Potosi formation is represented as resting disconformably upon the Bonne Terre or St. Joseph limestone of uncertain Cambric age, but probably at least Middle if not Upper Cambric. At the base of the Potosi is an edgewise conglomerate extending upward for about feet and followed by 100 feet of conglomerates and interbedded slates, the latter carrying several species of trilobites, brachiopods and an occasional Hyolithes primordialis. As was stated on p. 13, Beecher, who identified the fossils collected by Nason and who described the one eurypterid found, did not and perhaps could not state in just which layer Strabops occurred and whether it was found directly associated with the marine forms. From a study of the material in the Palaeontological Museum at Columbia University, I have found that the rock in which Strabops occurs is not of the same lithological character as is that in which the other fossils occur. The slab on which the counterpart of Strabops rests is roughly 3 x 9 x 12 inches in dimensions and contains no other organic remains. The limestone containing the trilobites is somewhat finer grained, differing little in color, but being made up of numerous cephala, pygidia and fragments of several genera of trilobites. The difference in faunal character between the two rocks is pronounced. The slab containing Strabops was not collected by the same person nor at the same time as were the other fossils, so that exact data probably will never be obtained. However, the precise association is of slight import. The alternation of limestone conglomerates and shales in the lower Potosi series indicates near-shore conditions of sedimentation, and the occurrence of the single specimen of a eurypterid, far from pointing to a marine habitat for this one individual, militates very strongly against such a mode of life. In all cases the occurrence of a single individual is one of the strong arguments against the assumption that the individual belongs to the fauna of the bed in which it is found. It is far more logical to assume that it has been brought there by some accident, for in Nature we do not find single individuals of any kind of animal in a region far removed from that occupied by other members of its family. Again, the only way to account for this occurrence is to assume that these eurypterids were living in the rivers of that time, and that this individual happened to be carried out into the shallow sea in which the Potosi limestone was being deposited. That the sea was shallow is indicated by the fine stratification of the rock as well as the paucity of the organic remains which are insufficient to have furnished the lime of which the formation is composed. This limestone like others of its kind seems to have been formed from the calcareous sand and mud carried by surcharged rivers coming from limestone regions into shallow seaborder basins.
The merostomes of the Stephen shale of British Columbia are not now recognized as eurypterids, but belong to a distinct order, that of the Limulava Walcott (Clarke and Ruedemann 39, 410). Hence their association with marine organisms may be disregarded.
3. It may hot be quite so clear that the occurrence of a fairly large fauna of eurypterids in a bad state of preservation, but associated with fossils of marine origin, in no wise indicates that the merostomes were marine. The Siluric fauna of Bohemia is one of the best illustrations of this class, and I shall consider it in detail.
Barrande's work on the faunas of the Palæozoic rocks of Bohemia has conclusively shown that the trilobites and other crustacea, as well as the eurypterids reached their acme in numbers in the Siluric, constituting the third fauna E. The upper part of this showed a far more prolific development of life than did the lower, as is readily brought out by the following figures. In the Lower Siluric (E e1) Barrande records sixteen species of trilobites and ten species of other arthropods among which he includes phyllopods, ostracods, eurypterids and cirripedes; for the Upper Siluric (E e2) the corresponding figures are 82 and 24, making a total of crustacea (and eurypterids) for the Lower Siluric of 26, for the Upper 106. Furthermore, the crustacea, though represented by so many species were not the dominant forms of life, for the Siluric, especially the upper part, marked the period of greatest development of the cephalopods which were represented by 665 species. As I stated in an earlier part of the paper, Barrande does not give horizons of smaller taxonomic value than his "bands" which correspond to the first subdivision of the periods, and it is therefore impossible even to approach the niceties of correlation which can be attained in America; one cannot determine the precise level even within several hundred feet for any particular occurrence. However, there is no reason to doubt that all of the Siluric of Bohemia was marine. Considering the nature of the fauna of that period and the number of species which Barrande was able to describe even so early as 1852, his explanation for the fragmentary character of the eurypterids, as due to their having been the food of the cephalopods, seems inadequate. If the trilobites were able to live in the same sea with cephalopods and escape unscathed, so that their remains were preserved in wonderful perfection, why should the eurypterids have been so voraciously attacked? It is doubtful if the eurypterids were of so different an internal nature from the trilobites that they should have been more palatable, nor were their exoskeletons more fragile. In the Siluric sea 814 species of cephalopods are known to have existed, as compared to 97 species of trilobites. Thus there were eight or nine species of Cephalopods to each species of trilobite, while the number of the individuals of the former vastly exceeded that of the latter. Surely in the great struggle for existence which was taking place, the cephalopods, if they fed upon crustaceous animals at all, would scarcely have used such nice selection so that the eurypterids alone were consumed, while the trilobites continued to flourish.
Of a certainty, some more rational explanation must be sought. This occurrence in Bohemia is one of the rare ones in the Siluric in which the eurypterids are found associated with an abundant and unquestionable marine fauna. Yet the facts, that no complete individual has been found, that even the fragments are of so uncertain a character that some which at first were supposed to belong to separate species have with more study been found to belong to the same species, and finally that the eurypterids, of all the myriad organisms which lived in that sea, should have been broken to fragments of which only a few are found—these facts will not admit of explanation on the ground that the eurypterids lived in the sea. They must have lived in some other aqueous realm besides the sea, and one is again led to the conclusion that they must have lived in the rivers. The facts of migration and the relations of the Bohemian forms to those in other parts of the world strongly support this conclusion. (See below chapter V).
- ↑ Lipalian is a term proposed by Walcott in 1899 "for the era of unknown marine sedimentation between the adjustment of pelagic life to littoral conditions and the appearance of the Lower Cambrian fauna. It represents the period between the formation of the Algonkian continents and the earliest encroachment of the Lower Cambrian sea" (290, 82).
- ↑ The Belt terrane is considered by Professor Grabau as representing a pre-Cambric Palaezoic terrestrial deposit lying above the true Algonkian.
- ↑ Professor Grabau has discussed this subject very fully in his lectures, with especial reference to the Graptolite beds of Europe. He expects to publish soon on this subject.
- ↑ The lithogenesis of this formation has been discussed in such detail by Grabau (Early Palæozoic delta deposits) that it is unnecessary to give more than a summary here.
- ↑ Professor Grabau has recently voiced the opinion, that the Pittsford shales and Shawangunk conglomerate are better considered as the closing deposits of the Guelph period.
- ↑ The distance would, of course, be much greater were the folds eliminated.
- ↑ Clarke and Ruedemann have placed Drepanopterus as a subgenus of Stylonurus.
- ↑ Laurie originally spelt the name Bembycosoma and the species of Drepanopterus, bembycoides, but in the corrigenda to his 1900 paper, p. 500, (147) he called attention to the proper form Bembicosoma and bembicoides, a change which has not been noted by later authors.
- ↑ He had been led to look for this fauna because he had noticed in the Dorpat Museum certain fine specimens which had been sent in from Arensburg, southeast Oesel, by Oberlehrer Werner, who had knocked them out of loose blocks of building stone. (Nieszkowski, 197, p. 303.)
- ↑ It is true that pioneer work on the mainland had been done by M. v. Engelhardt and E. Ulprecht, the results being embodied in a paper entitled "Umrisz der Felsstructur Ehstlands und Livlands" in Karsten's Archiv für Min. Geogn. Bergbau u. Hüttenk. for 1830, but the paper does not touch on Oesel. Similarly in the Geology of Russia by Murchison, de Verneuil and Keyserling Oesel is passed over in a few sentences.
- ↑ This species was not collected by Schmidt but was described by him from a specimen from Volborth's collections.
- ↑ It does not appear to me necessary to take up in detail the discussion of the occurrence of the Pterygotus marl of Gotland, since the conditions there are identical with those of Oesel. The marl overlies beds with a Wenlock fauna, and is succeeded by beds with an Upper Ludlow fauna. The physical evidence of the break between the two series is marked throughout the island. This is fully discussed in a forthcoming paper by Professor Grabau.
- ↑ Journ. Acad. Nat. Sci., Philadelphia, vi, 2d series (1860–69), p. 207.
- ↑ The bright red Vernon shale (Salinan) has shown on analysis only 2.25 per cent of ferric oxide and 0.75 per cent of ferrous oxide.
- ↑ The significance of this fauna has already been discussed in chapter III, p. 92, and the other aspects will be considered below, p. 247, et seq.